A Re-Upload of High Effort Reddit Posts
Somehow, I got banned on reddit. I’ve appealed. I don’t know what it was for, but false positives happen.
My best guess is that someone reported my Sovereign launch post as having being against the terms of service or something. I was able to confirm that reddit’s automod removed that post initially and a moderator was able to help me out by re-enabling it. Just grasping at straws here.
I’ve had the account for 9 years, and I’ve posted a lot of high-effort stuff. Currently, no one but me can view it, and I think that’s a shame. I’ll try recapture the highlights here in chronological order.
On The Math of Pathfinder 2e Encounter Balance
Say that you plan on leveling up from level 1 to level 20 using the same party of 4 players gaining XP from nothing but moderate (80 xp) encounters. That would be 12.5 encounters per level times 20 levels for a total of 250 encounters.
Taking a massive simplification here, if the probability that a party doesn't TPK (ignoring complications like a character dying but the party still wins) is X, then the chance that the party makes it through level 1-20 is X^250.
If "moderate" encounters were actually difficult, even in pf2e, then a campaign would be impossible.
Say that you want to make it so that even with proper tactics there's still a real chance of failure - say 5%. That means that in just three levels, the party's survival rate has gone down to 15%. the campaign survival rate has gone down to 0.0003%.
What about "severe" encounters? Say you plan on giving your party one of those per level, so they'll have to clear 20 for a full campaign. If each of those severe encounters has a 10% failure rate, there's only a 12% chance the party completes the encounters.
So here's the game design paradox. Even in systems with "integrity", the game needs to provide the illusion of danger without there being actual danger. If every encounter actually had a non-negligible risk to the party (even 5%), then the party would just get blown out by variance over time.
It sort of sucks once this veil has been pulled back.
On Pathfinder 2e Complexity and Trap Options
A lot of the hidden complexity of pf2e is tied into the trait system, and that's where a lot of the "gotchas" come from. Here are some examples:
Calm Emotions seems like a super good spell until you check the trait list and read about what the Incapacitation trait does.
Likewise, you can't finish off a low-health enemy and then Sudden Charge to another target because it has the Open trait.
It works the other way. It would seem ridiculous that a Ochre Jelly would be susceptible to Trip. It has a long list of immunities, but that list doesn't include immunity to prone, flatfoot, or tripping. Further, even though it's immune to critical hits, weapons have critical specialization effects and some weapons like those in the Flail group automatically cause the victim of crits to get knocked prone. Since the Ochre Jelly has such a low AC relative to its level, it gets knocked prone by flails (like the extremely popular gnomish flickmace) over and over.
Lots of weird stuff in the corners like that.
A lot of the spell balance seems extremely off to me, from a numbers perspective. Take, for instance, Field of Life vs 6th level Heal.
Field of Life heals for a paltry 1d8 (4.5 average) HP per turn, and characters capable of casting 6th level spells will level 11. Fighters will have ~150+ HP by then, so they're getting healed for 3% of their health every turn, and it's costing the caster an action every turn. Enemies can simply walk into the radius to start getting healed for free themselves.
Compare this to the 3-action version of 6-level Heal. You cast it once and it heals 6d8 (27 average). No need to spend further actions. You can also flexibly cast this as a single target heal for 6d8 + 48 (75 average). Much much much more useful.
The book is full of extremely niche (read: trap) options like this.
From the community management side, I think paizo does a worse job of answering rules questions than the 5e folks do. Mike Mearl is honestly great (or used to be, I don't check anymore) at answering/compiling answers about rules clarifications on twitter, and then posting a big FAQ. Paizo has been less stellar.
But compared to Kromm and the GURPS community the difference is night and day. I don't know how Kromm keeps it up - it's amazing how much patience that man has to answer so many precise questions with so much detail.
For instance, try to figure out how Wall of Stone works. Good luck!
How to Perform Breadth-First Research
It's far more effective with almost everything you want to learn to do what's called a "breadth-first" search rather than a "depth-first" search. If you're reading paper about DNA on Wikipedia and come across the paragraph:
Deoxyribonucleic acid is a molecule composed of two polynucleotide chains that coil around each other to form a double helix carrying genetic instructions for the development, functioning, growth and reproduction of all known organisms and many viruses. DNA and ribonucleic acid (RNA) are nucleic acids. Alongside proteins, lipids and complex carbohydrates (polysaccharides), nucleic acids are one of the four major types of macromolecules that are essential for all known forms of life.
and you're relatively new to chemistry, you might have a bunch of questions. What are polynucleotide chains? What is a molecule? What is a double helix? What are genetic instructions? What are macromolecules?
If you do a breadth-first search, you might note down that you don't know these things as research-candidates, look up just enough to be able to piece together the meaning of the sentence, and then try to understand the sentence you're reading before getting lost in the weeds.
If you do a depth-first search, you hit "polynucleotide chains", and then that brings you to
A polynucleotide molecule is a biopolymer composed of 13 or more nucleotide monomers covalently bonded in a chain. DNA (deoxyribonucleic acid) and RNA (ribonucleic acid) are examples of polynucleotides with distinct biological function. The prefix poly comes from the ancient Greek πολυς (polys, many). DNA consists of two chains of polynucleotides, with each chain in the form of a helix (like a spiral staircase).
And then you're like "what's a biopolymer", before you can even finish reading the article about polynucleotides, so you start reading about biopolymers and you hit
Biopolymers are natural polymers produced by the cells of living organisms. Biopolymers consist of monomeric units that are covalently bonded to form larger molecules. There are three main classes of biopolymers, classified according to the monomers used and the structure of the biopolymer formed: polynucleotides, polypeptides, and polysaccharides.
And then you're like "okay, they're natural polymers but what are polymers and what are monomeric units" and so on. Eventually, you find yourself reading about quarks or the axiom of truth and you're like "wait a second I just wanted to know about DNA how did I get here?". Not efficient at all.
The breadth-first search uncovers what you don't know, and then you decide later if you want/need to know it. Then you read about that topic and repeat the process. Applied to the topic of learning pf2e here's how I think my learning was structured:
What's the gameplay loop of pathfinder? Describe scene, players make decisions, potentially roll dice, describe new scene.
What sorts of scenes am I describing? Combat scenes, exploration scenes, RP scenes, downtime scenes, scene transitions
When dice are potentially rolled, how? According to scene. Combat scenes use combat rules, the other scenes mostly use skill checks.
How does each scene type work? etc
How is combat structured? etc
How does a skill check work? Roll skill on Character Sheet + d20 vs DM's secret number
How do I make a character? etc
You end up creating a structured map of the information that you fill in at your own pace. I think it's important to skim through the book at least once so you know where to look for answers and how to structure your investigation.
Folks do this sort of thing fully mentally, with simple notes, or with something as involved as https://obsidian.md/
On The Gradient of Pay To Win (P2W) In Games
Rather than thinking in in black and white, let's think in terms of "how much advantage can a player with a lot of IRL money buy?" This gives us a much better way to reason about the gradient of "pay to win".
In a game like rocket league, what can you actually buy? The in-game purchases do absolutely nothing, they're fully cosmetic. But, if you're wealthy, you can buy better hardware, coaching, and potentially rating boosts. If "winning" is reaching masters league, it's way easier to get boosted there than to get good enough to get there yourself. If "winning" if being good enough to reach masters league on your own, it's way easier to reach that skill level by paying a coach to train you than coming up with your own training routine. It's easier to do both things if you have 240fps than if you're playing on 60fps (which is expensive).
In a game like WoW, all of the stuff you can do in rocket league you can still do, except in addition to all of the above you can also convert your IRL dollars into in-game currency, and then trade in-game currency for a faster path to better gear than the players who don't pay.
What's the fastest path to being in a cutting edge WoW guild? Pay for gear boosts and raider.io score, and pay a coach to teach you the fights so that when you do your trial you're good enough to pass it. You could go from not knowing how to play to being a competent fully geared raider orders of magnitude faster than someone else who isn't willing to pay. The person who refuses to open their wallet is farming for gold to buy boosts (which cuts into their practice time), or even worse, is trying to do all of their runs legitimately, which looks like standing around in oribos hoping a group accepts them (which cuts into their practice time).
I think with that in mind, it's easy to say that in wow you can buy more advantage than you can buy in rocket league. Or, phrased in another way, WoW is "more P2W" than rocket league.
At the far end of the spectrum, you have Archeage!
On MMO Quest Design
Before we can talk about quest design, we have to understand why we even level up in the first place, which is rooted in character power vs player power. Typically MMO's provide a number of different ways that the character can become more powerful independent of the player getting better at the game - character level, gear, skill points, borrowed power systems, etc.
From a game design standpoint, we often think about "character level" as the anchor for the rest of the systems. Gear typically has minimum level requirements but levels don't typically have minimum gear requirements. You gain new abilities when you reach a certain level, and start suddenly being able to fight more difficult foes, and earn less experience for defeating the same foes that used to give you trouble. Narratively, your character (not you) is experiencing character growth.
Narratively, deeds done (which include quests and defeating enemies) explain why your character is growing. It makes intuitive sense that someone who has saved the town from the defias bandits, rescued westfall from the gnoll incursion, and stopped redridge from getting overrun from wyrmkin would be stronger than a neophyte adventurer.
Typically the narrative structure lies somewhere in the gradient between the western TTRPG idea of small-time adventurers traveling from town to town, dealing with local problems and growing in reputation and importance until they're eventually combatting existential threats (vanilla wow) and the JRPG model of the-chosen-one slowly filling the shoes of their destiny as they progress through their (linear) destiny (ffxiv).
Back on track: It is extremely important to both of these narratives that the character progression seems in accordance with deeds done. It doesn't feel like a stretch when we grind 1000 difficult mobs and then finally gain a level and unlock a new small stat boost and an additional ability that helps us be slightly more efficient in combat (some sort of non-flashy cleave, for instance).
It feels extremely discordant when we're following the main quest in FFXIV and we're delivering notes, messages, and wine bottles from one NPC to another, teleporting constantly, not killing things, and have gained 4 levels in the last 2 hours and unlocked 2 new abilities that cause our characters to turbo-anime the next bad-guy we have to fight. We used to not do that, why did we have a power spike? From delivering messages? Why don't other people deliver messages if it confers this kind of power? Is our chosen-one-power that we gain extreme combat prowess from being a messenger?
Once you make sure you're not breaking immersion by lining up character progression with deeds-done, you want to make sure that your game knows where it wants to be on the gradient between ttrpg ←→ jrpg and sticks to it, and that in either case players feel like they have agency. If you've made a jrpg linear main quest, and then you stick a google-maps navigator into your game to tell people where to go all the time, what exactly are they doing? You've solved the puzzle of "where should I go" for them, and as I've discussed the actual games tend to be easy. Much better to obscure your linear path via something like hollow knight's world design where there are multiple ways to "discover" the only path to where they needed to go, so it feels like they found it naturally.
If you want your players to have actual agency, make sure that when they finish up a quest hub or grinding area or whatever, you give them multiple breadcrumb quests (or no breadcrumb quests), so that they choose where they want to go.
Further, the ways that players have to actually interact with the game world in a MMO tend to be pretty constrained. You can click-to-interact, kill stuff, or go-somewhere and that's about it. If you allow players to permanently/temporarily change the state of the game (destroy a bridge, repair a building, etc), then all of the other players attempting to do the quest at the same time either need to see their own "phase" of the world (which is extremely problematic), or everyone else gets locked out by the first person who does it until it resets somehow (which is weird). This heavily constrains the sorts of stories you can tell through questing.
Questing systems like New World are mindnumbingly transparent about the fact that all the quests are click-things or kill stuff. Runescape is still click-things or kill stuff, but they try extremely hard to dress up the puzzle of "click things and kill stuff" via the metroidvania concept of locks and keys. That effort translates into a much more satisfying questing experience (and it's less immersion-breaking when our characters can do cool stuff as a result of making it through the ordeals).
Finally, players will optimize their own fun right out of the game if you let them, and XP/h is king. You need to make sure that whatever you think that players think is the "most fun thing" needs to also be the best XP/h. Your quest might be flavorful and emotionally resonant, but if the players have to choose between doing it or just killing nearby crabs for 1.5x xp/h because you didn't balance it properly (spawn rates, xp rewards, travel time, etc), then the players will kill crabs and complain about it.
Tangentially related: A lot of games include both a leveling system and a gearing system (and other vertical progression systems), but the gearing system only actually becomes relevant when you hit max level. For instance, if you're playing WoW (especially retail), how much do you actually care about your gear as you level up. Do you take a break from grinding levels in order to try to find better gear? Of course not! That gear will quickly become obsolete. It would be silly to run the deadmines a bunch of times for best-in-slot level 18 gear when you could spend that time just hitting level 25 faster.
This creates this weird dynamic where the game keeps giving you all these items that you don't actually care about the whole time that you're leveling. Like yeah, you check to see if the item is an improvement, and if it is you put it on, but you're not checking deeply. You don't go out of your way for better items. Then when we're done leveling, all we care about is gear, and if you're a dedicated MMO player, you care about gear for a far longer percentage of your total game-time than you care about experience. It's a startling transition for new players (who might be inclined to care about gear while they're leveling, or taught to not care about gear once they're done).
Here's my favorite solution: make both matter. Make "soft gear checkpoints" where level 21 (or whatever) monsters have significantly more health / do more damage than level 20s. If you try to go fight level 21 stuff with level 17 trash gear you get dunked on. Now, in order to progress efficiently, you need to go out of your way to find/buy/make best-in-slot / near-best-in-slot items to fight these more dangerous monsters. Make players interact with the gearing system early on.
On New World’s Launch and Game Design
When we set out to make a MMORPG, there's a few things we want to accomplish:
It should feel massive-multiplayer-online, meaning that you get a sense of online community
It should feel like a RPG, meaning that you get to role-play a character
The world, in order to feel massive, should offer opportunities to explore
Part of developing a character is having some sort of vertical progression
Tying that vague outline back into your feedback, we see how new world falls short. In making all of the landmarks too samey, they also make them predictable, and so future exploration feels pointless. Even though there might be hundreds of landmarks left unexplored, you don't feel compelled to explore them because you have a sense of "seen 1, seen them all". The same can be said for the vertical progression of our characters. My character didn't feel especially stronger playing new world at level 45 than it did at level 15. I didn't have new abilities, I was fighting the level 45 skeletons doing 15% of their health with a m1 and it was the same gameplay as fighting level 15 skeletons at level 15 doing 15% of their health with a m1. The only sense of vertical progression was that I had a higher abstract "gearscore" and "character level", backed by hundreds of repetitive "deeds done" - I had done thousands of pointless samey quests.
In other games, you slowly unlock new abilities, your gear progresses visually, the scope of your deeds increases (first you deal with small-town ruffians in elwyn forest, then you save westfall from the larger vancleef bandit plot, then you save duskwood from the werewolves, undead, and stitches, then...) and so the vertical progress of your character's power makes sense.
I think all of these problems are more or less dwarfed by the larger problem new world has though, which is that it doesn't have a coherent vision for what to do with end-game players. Everything you described are valid criticisms for players in their first 300 hours. A MMO needs to support someone who wants to play for 5000 hours minimum, in my opinion. By the time you're 300 hours deep in a MMO, you will have explored the world, found the factions, leveled your character to max, etc, but you're only 15% done with your playtime. What's left is how the developers handle
the player economy
fostering pvp and pvp balance
releasing new high-end pve challenges
New world has a leaky-faucet problem. Gear doesn't go anywhere so once players get the best stuff, the best stuff keeps dropping and is worthless. Gold stockpiles and is worthless. Players arbitrarily decide that they want to raise their crafting number and so they buy stuff to craft pointlessly and make it, but the stuff they make is worthless, further contributing to the problem.
The content that players are actually interested in: fighting each other and having ever-more-difficult-pve (think raids and m+ in wow) doesn't create a sink on the economy so all the stuff that players are gathering doesn't go anywhere. If the actual gameplay is also sub-par: the incentivized pvp isn't especially interesting or it's unfair, laggy, shallow, etc, and the pve is also too-easy or uninspired, or they don't release new challenging pve often enough for players to consume, do you even have an endgame?
I have a different set of concern for ashes. It seems like they have a pretty good handle on the end-game loop and sinks+faucets, and I don't really care about the questing/leveling experience (because again, only a small percentage of your playtime is in the leveling phase). What I'm really worried about is that:
Since players can buy everything they need (it's an open market game), and since players are incentivized to specialize economically, what stops the game from degenerating into the player constantly performing their specialization until they are rich enough to buy whatever they want? IE: the game offers tons of different activities, but as a mining specialist, it's inefficient to do anything but mine until I have enough money to buy my full best-in-slot gear.
The game looks to encourage 8-man parties, and further wants those parties to be composed of exactly 1 of each of the primary archetypes (fighter, tank, rogue, ranger, mage, summoner, cleric, bard) otherwise your party will be inefficient (missing buffs / synergies) and will be unable to pass through specifically designed dungeon challenges (only the mage can disable the illusion, only the rogue can pick the lock, etc). I'm afraid this will lead to extremely annoying group-creation dynamics where you're always spamming chat for a bard or whatever, and your real-life friends better all agree to play different classes, and hopefully, if you make a new friend in the future they're a different class than you and all of your current friends.
The game highly encourages and rewards PvP (you can loot people after killing them), but at the same time, looks roughy impossible to balance for PvP. 1v1s are going to be wildly imbalanced where there will be matchups you lose even though you're way better than your opponent. I imagine that there will be 2v2 or 3v3 comps that are just flat-out better than other ones, and that because players have so much build customization intrepid will have little-to-no control over what this is. To make this worse, switching to whatever the best is will take a massive time commitment, so I guess enjoy dying for the next few months if you happen to be underpowered in the current patch.
How does ashes plan to have actually difficult PvE bosses? You can't have a 10 minute long bossfight that makes the players juggle 15 mechanics flawlessly if you can get stunned by a stealth rogue while the guy with the bomb is supposed to be running it into the teleporter. It took the best WoW guild 700 consecutive wipes on mythic jaina before they got their first kill.
On Balance in MMOs
The very concept of 8-man balance with "rock paper scissors 1v1s" is pretty batty / smoke-and-mirrors to me.
First, try to define what balance is, and what that word means for satisfying play, especially in a MMO context. Say that you have a game where there are three classes - mage, fighter, and rogue. Mage > fighter, fighter > rogue, and rogue > mage, and the population is split up into these three classes equally. Say that when an equally skilled mage fights a fighter, there's a 90% chance they win (rps "balance"), and the same for the other matchups.
If we define "balance" as "no class has a higher average winrate than any other class", then we did it. I argue, though, that that's a terrible metric to use, because the game feels horrible to play. Matchup rather than skill is the determining factor in most battles, and that's deeply unsatisfying. As a rogue, I should run from fighters (unsatisfying), and get free wins against mages (unsatisfying), and actually get to test my skill against other rogues (satisfying).
If we define balance as "the game is more balanced the lower the standard deviation from even matchup scores we have", then we have wonderfully satisfying and balanced games. Say you have a game where you have a warlock, a ranger, and a paladin, each with very different play archetypes, but the 1v1 matchups between all three of those classes are even. If we have nothing but even matchups, then no matter who you encounter, you can battle them on even ground and it's the player (or character power differential) that decides the outcome.
Competitive 1v1 games strive toward the second one, because intentionally building for rps balance would be lunacy. Why would I play a game where when I show up to a tournament, how deep I go is almost exclusively determined by whether I face my counter in bracket? Why design a game where players auto-lose some large percentage of their battles because of the system?
The second intuition is that the more options you have the harder it is to balance. Starcraft 2 has 3 races: zerg, protoss, and terran. Mirror matches are automatically balanced, so we need to make sure that ZvP, ZvT, and TvP are all balanced and then we're done. If we were to add a 4th race, Gorloks, then we need to balance ZvG, PvG, and TvG, effectively doubling the number of relevant matchups for balance.
In ashes, we have 64 classes, each of which will presumably have multiple viable builds (if the marketing is to be believed). Say this number comes out to be ~3. That's 192 viable builds. That means that each build (which is the equivalent of a sc2 race: you don't play a class you play a build), has to worry about 192 matchups, and for the game to be balanced on a 1v1 level (192^2-192)/2=18,336 matchups to be balanced. If instead, the subclasses aren't viable, and there's only 3 viable builds per primary, then there are ~24 viable builds and (24^2-24)/2 = 276 matchups to be balanced.
The third intuition is that balancing around team play is exactly like balancing for 1v1s. Starcraft 2 has a competitive 2v2 scene, though it is smaller than the 1v1 scene. How many matchups are there? You have ZZvZZ, ZZvZP, ZZvZT, ZZvPT, ZZvPP, ZZvTT, ZPvZP, ZPvZT, ZPvPT, ZPvPP, ZPvTT, etc.
It doesn't make sense to ask "how good is zerg in 2v2?", it only makes sense to ask "how good is zerg+protoss in 2v2". Races aren't played in 2v2, teams are, and the teams are composed of races. You might then look to see if one race is part of more top compositions, but that's ultimately irrelevant. If the top team is Terran+Terran, and it's dominating the ladder, but terran doesn't show up in any of the other top teams, is terran bad? We're asking the wrong question. The only thing that matters is that terran+terran is good.
Given that balancing for team play is exactly like balancing for 1v1 play, we need to know how many teams there can be so we can figure out how many matchups there will be (since the more matchups there are, the harder it is to make sure they're all balanced).
Say that we know that each 8-man team needs 1 tank, 1 cleric, and 6 damage dealers. Say that there are only 3 viable builds per primary archetype. That's 3 options for tank, 3 options for healers, and 3•3•3•3•3•3 options for damage dealers, for a total of 3^8 = 6,561 teams. The number of matchups the ashes team would have to balance is ~21 million. If each class had 24 viable builds instead, or if the party compositions could be more flexible...
The conclusion here is that the number of matchups is completely intractable for any sort of balance to occur. They also want to make it costly to switch, as far as I can tell they don't just want you hopping around to whatever the meta primary archetype + secondary archetype + gear + augments combination (aka your build) is, so you're effectively at the mercy of the dev's balancing effort which looks to be hopeless as far as I can tell.
Pathfinder 2e Gripes
Here are my gripes:
The disconnect between actual danger and perceived danger. Characters will have ~10 encounters at every character level. If we define a "dangerous" encounter as one where there's a ~5% chance that a character dies, then we would expect a character death every two levels, and 10 characters deaths across every campaign. We have to create the feeling of danger (or else why play the fight out?) while hiding that there's a far less actual probability of death (the encounter is actually non-threatening, mathematically).
The lack of situational variation for most martials. After gaining a degree of system mastery, martials feel extremely flowcharty to me. Flank your enemy, hit them hard, use whatever third action you built your character for. I'm glossing over a bunch of complexity here (don't trip an enemy right before their turn, delay until right after their turn, etc), but once you've dialed in all that complexity, I find my table's martials executing the same 2-3 optimal plans in basically every turn and every combat.
Trap option spells. Some of the spells are really bad, and as they continue to make more, it gets harder and harder to onboard new casters via 2e.aonrpd. They're shown a wealth of traps they have to avoid to find the good ones.
Inflexible-ish builds. Want to play a high-int, low-strength barbarian? Your character is going to suck. See this for what build flexibility could be like.
Spell descriptions are vague / confusing and the designers refuse to clarify. How does wall of stone work? Can you use it to make a box? If it has to be aligned on a grid, how can you use it to make stairs? Can it be layered?
How does control water work? You can raise/lower the level of a 50x50' area, but what happens if the pond is bigger than that. You lower a little block of the pond by 10', does other water rush to fill the vacuum? If you raise the water above ground level, does it stay as a weird water pillar, or does it even out into the pond/ground? If it goes into the ground, how much water? Do get 50x50x10 cubic feet of water spillage and the level doesn't raise at all? Can you control water in an area with no water? If not, does spit count as water?
Incoherent dungeons. This one is more about the genre, but also about how tightly balanced pf2e is. Say you're running a paizo official adventure path and you're delving into a dungeon. It's been my experience that most of the dungeons:
expect you to rest after every fight to heal
potentially expects you to long rest
expect you to clear it room by room
assumes the GM makes basically everything fight to the death on sight
doesn't want the NPCs running to group up and gang up on the players to create extreme/tpk-difficulty encounters
For instance, say that you were adventurers going to a bandit lair. Your party busts through the front and get confronted by 4-5 bandit thugs which are aptly difficult for you. They all fight to the death. In the next room right behind the door is another group of 4-5 bandit thugs, maybe with a thug mage or thug dog or ogre or whatever, and so on for 4 or 5 rooms until you get to the big bandit boss.
The way this should play out is that unless you somehow manage to subdue all of the original bandits both instantly or silently, the rest of the bandits should know immediately, alert the rest of them, and you'd have 35 bandits screwing up your 10 refocus session in a heartbeat. Knowing this, you might choose to siege the bandits, smoke them out, cave them in, whatever. But then it wouldn't be a dungeon romp. So, we have tightly balanced encounters and the enemies play by those rules too. The second-to-last encounter grouping up with the boss encounter creates a TPK encounter, so that can't ever happen or the dungeon is bricked.
Consumables suck. Say I get a truth potion as a reward at level 6 (it's a 6th level item after all). That just ate up 46g of the treasure budget, and now I can either sell it for 23g, or I can try to find a situation where I can convince someone to drink it (or force them to drink it), at which point they get a DC 19 Will save or have to tell the truth. 6th level creatures have Wills in the ~+13 region, so they're succeeding that save 75% of the time. If they succeed, they can act like they failed to feed you false info. If you don't find someone to feed it to soon, the enemy's Will saves will only go up, and the item will become less useful. They're super situational, and players forget to use them even in those situations. They compete with gold to buy the potency/striking runes that the game's math is balanced around.
Recall knowledge needs to be systematized. This is a decent start.
On Pathfinder (and most TTRPG) Combat Being A Separate Game
https://www.reddit.com/r/Pathfinder2e/comments/xxbabz/how_to_say_yes_and_with_a_difficult_player/
There's something important here that I've written about a few times that I've taken to calling the "Two Games Principle".
Perhaps easier to illustrate by watching a few minutes of Pokemon Blue.
Notice how the player character is exploring the forest, walking around obstacles, potentially interacting with NPCs, hopping over logs, walking through tall grass, and then BAM! at 3:01:17 the screen goes dark. The music changes into something hectic. The ui completely changes into a battle UI. The way the whole game is played completely changes into a battle simulator rather than an explroation/interaction simulator. "A wild PIDGEY appeared!. What do you want to do?"
FIGHT
ITEM
PKMN
RUN
Your options are extremely constrained. If you choose fight, that now opens up another limited set of options, all of which have very predictable results.
In my view, pathfinder (and ttrpgs in general) can be played this way, and if everyone signs up to do so on a social contract level, they work better. They tend have hundreds of pages of in-depth rules to support this exact mode of play. What this requires is that you interpret the rules exactly. If you crit an ooze with a flail-type weapon, it gets knocked prone. Prone says you're lying on the ground. You are flat-footed and take a –2 circumstance penalty to attack rolls. The only move actions you can use while you're prone are Crawl and Stand. Flat-footed says you take a –2 circumstance penalty to AC.
Is any of this fictionally coherent with an ooze? No! Oozes don't have feet. How can they be flat-footed. How can oozes "fall" or "be prone"? Doesn't matter. Each condition comes with mechanical effects that interact with the other mechanical effects. The ooze has -2 to attack and -2 to AC and can only Crawl (5ft move) until it uses an action to Stand (triggers an attack of opportunity). It doesn't matter what the flavor is, these are the mechanics. What's important is that the Ooze isn't immune to prone in its statblock.
Same deal with Light. The spell doesn't dazzle in encounter mode. If it did, it would say so. Everything, and I mean everything, does exactly what it says on the tin, no exceptions.
This allows players to reason about the game and make future plans without the sense that they're playing Calvinball. It allows the GM to be fair and consistent when the PCs' lives are on the line. It encourages and rewards system mastery for the folks who seek that (since now there's a coherent system to master).
So you have your (at least) two games. Most of the time (at some tables), you're modeling the fiction of the world. The effect you make with light can get dimmer or brighter on a gradient, and you can send it into someone's eyes and temporarily blind them as a prank. You might be in a dungeon trying to avoid traps, or sneaking around a city charming guards and dukes. That's the first game.
The second game is when the music changes and the GM calls for initiative to be rolled. Suddenly now we're playing a wargame, and everything that was really wishywashy and whimsical gets really precise. Now everything does exactly what it says on the tin. Now, rather than telling a story, and interpreting that using mechanics (fiction first), we use mechanics, and tell a story based on what happened (mechanics first).
A Litany of Pathfinder 2e Complaints
Pathfinder 2e does a bunch of stuff well! For instance, rather than tracking weight directly, it uses something really close to item slots. Monsters, sort of like in 13th age, tend to be interesting rather than bags of hit points. It has a huge bestiary and a huge magic item list (which offloads tons of creative work from the GM). The encounter builder actually works (the amount of indie darlings with no concept of how to predict fights ahead of time is staggering). The rarity, item level, and town level system is wonderful. Buying, selling, and identifying items is painless. The rules are all free and available online (this is gigantic).
But, here's a list of faults, as requested
Recall Knowledge / Creature Identification, read as written is a bunch of nonsense. I wrote this homebrew to fix it.
The spell lists are filled to the brim with trap options. Others will tell you that you can't make a bad character. You can if you're a caster and you pick a bunch of bad but flavorful spells. Compare Heal or Magic Weapon with Spirit Link. There are tons of such options at every level in every spellcasting tradition.
Lots of spells have difficult-to-parse outcomes. For instance, what does Control Water do? What happens if you control water in an empty room? Do you need a source of water? Is spit enough? If the body of water is bigger than the area you're controlling, does the water hold shape like a pillar/divot, or does it flow back into the broader water? Does this generate an effectively infinite amount of water if you do it on top of a mountain? All unclear. For another example, how does Wall of Stone work? Can you layer the walls? Can you build cubes?
It's unclear whether the mechanics take precedence or the fiction. For example, can you Trip an Ochre Jelly? It has this huge list of immunities, but prone isn't on there (and it is on other creatures).
The game is very unsupportive of sandbox "combat as war" style play, leaning very hard into "combat as sport". If you've ever played a game like Divinity: Original Sin or even the original World of Warcraft (or MMOs before it), the idea is that the world does not adjust itself to the characters. The enemies are whatever level they are, and the threat level is coherent with the environment, not the characters. The characters choose the environments they go to based on their self-assessment of power/ambition or risk/reward rather than the challenges being crafted to make them feel heroic. In PF2e, because of how the 4 degrees of success work, and how quickly power scales (both for players and monsters), the band of appropriate challenges is really small. Characters will dunk on anything below them, and will get dunked on by anything too far above them, with very little room for negotiation.
The game has very little in the way of resource management. Create Water is a first level spell (don't really have to worry about water). Light is a cantrip (don't have to worry about torches). There's infinite HP healing after combat with trivial (required) investment in the medicine skill (don't have to worry about HP). That means that the only actual dwindling resource is spell slots, and those recharge daily. Threats must either get the players to spend spellslots or be potentially lethal, or they were just there for entertainment value. The 15 minute adventuring day is in full effect.
The bestiary does not include information about #appearing, or what sort of treasure a monster usually has, so in a sandbox campaign the GM is in charge of figuring out how large a roving pack of bugbears should be, or how much loot they should have on them, and how much loot should be in their horde back home. Contrast that with the first version of D&D. It says that wandering bugbears come in packs of 5d4, and their hordes have Treasure Type B.
The published modules and adventure paths are written not to be easily run at-the-table, but instead to be entertaining, since a large part of the consumer base buys adventures like magazines rather than as a play supplement. See these review standards for some sane ideas about what makes a good adventure. Here's an example of a conversion that makes it less fun to read like a book, but much more playable. Further work to assist the GM is doing stuff like including timelines and "spoilers" at the beginning of the adventure so the GM knows the structure of the plot, rather than having to "read to find out what happens". Or, knowing about all of the key players for the entire adventure path up front. This makes it less fun to read as a novel so Paizo doesn't do it. Compare a Paizo AP with, for example, Three Days to Kill or The Complex of Zombies in terms of how easy they are to run at-the-table.
The game doesn't include game structures for running a dungeon crawl. For instance, Worlds Without Number and Old School Essentials both have explicit timekeeping. Time happens in 10 minute increments (a dungeon turn). Depending on how hostile the dungeon is, you roll a d6 every turn, every other turn, and so on, and on a result of 1 or 2 you get a wandering encounter. Torches go out every 6 turns. Picking a lock takes a turn. Searching a room takes a turn. This makes the dungeon feel alive and inhabited without being stocked with a bunch of static encounters waiting for the PCs to come barging in.
The game has a number of Disassociated Mechanics. Every time the player reasons about hero points, they become un-immersed in the character. "I can't use my hero-ness right now or I won't have it for later unless I do something very much true to myself" is a totally incoherent thought for a character to have. Classes/ancestries tend to have per-day abilities that aren't at all explained why they are limited to once a day.
The crafting system is really bad. There are enough other comments talking about this to not elaborate.
Stealth. As written, stealth feels not that useful. If a party wants to avoid multiple fights in order to steal some sort of treasure or sneak into a complex, that means that each player has to roll stealth to get past each guard. This quickly rolls to failure. You inevitably get caught somewhere in the middle of the complex and now you're flanked between the two sides of the enemy forces, and it would have been smarter to just go in from the front. If you were fighting anyway, it would have been more efficient to just spend the points on stealth proficiency on something else, which is what ends up happening next campaign.
The way that the XP and loot incentives are set up, players are incentivized to kill everything. Every encounter cleared is worth XP and potentially drops valuable loot. That means if you have some sort of complicated dungeon complex like, say, the abomination vaults and tons of ways to make it from the first floor to the second floor, none of those actually feel like choices. Rather, optimal is to clear the entire floor, extracting as much XP and loot from the whole floor as possible before moving down to harder challenges. This has been a problem since D&D 3.5. Instead, we want to either reward XP for accomplishing things (like looting treasure or finding the next floor), and make sure that all of the random monsters/rooms don't have treasure, and that they're in concentrated treasure hordes only (goals).
The skill feat system is a mess. Some skill feats are combat useful (medicine, intimidation, athletics, bon mot), and the rest feel very iffy.
On the Downside of Tight Encounter Balance
Consider a hypothetical system where level 1 characters have a +10 hit bonus, and level 1 enemies have 20 AC, so you hit on a 10 or better.
Then, when you level up you gain 100 +hit and 100 AC, so level 2 characters have 120 AC, and +110 to hit, and level 2 enemies have roughly the same.
Level 2 stuff would never miss level 1 stuff, and level 1 stuff could never hit level 2 stuff. This is exceptionally tight math, it's just that now you can only effectively fight things that are exactly your level, otherwise the fight is either trivially easy or crushingly difficult.
If you do the same thing again, but raise instead raise the stats by 3 every level: level 1 characters have a +10 hit bonus and around 20 AC, level 2 characters have a +13 hit bonus and around 23 AC, level 3 characters have +16 and 26 AC, then we can see how this is still pretty suffocating for sandbox style gameplay. If something is 2ish levels higher than you are, you're looking at hitting on a 16, and it's hitting you on a 4.
The more of these systems are tuned strongly to level, the harder it is to have challenges that just exist, rather than crafting encounters for PCs.
For instance! In a OSR-style sandbox game, like Worlds Without Numbers, maybe a 3rd-level party has a 90% chance of beating a 2nd-level encounter, 80% chance of being a 3rd-level, 65% chance of being a 4th-level encounter, 50% chance of being a 5th level encounter, 35% chance of beating a 6th-level encounter, and a 20% chance of being a 7th level encounter (just making numbers up here). Whereas in "tighter math" games, a 3rd-level party probably has a 99% chance of beating a 2nd-level encounter, a 95% chance of beating a 3rd-level encounter, a 70% chance of beating a 4th-level encounter, a 30% chance of beating a 5th-level encounter, and a 1% chance of beating a 6th-level encounter.
The probabilities are much more focused on the challenges being exactly at the characters level, since that's where the "tight math" shines. Good for being able to create crafted encounters, bad for letting the characters wander a dangerous world that doesn't adjust around them.
A Litany of DFRPG (GURPS) Complaints
https://www.reddit.com/r/gurps/comments/10aambp/given_the_wotc_drama_is_this_a_time_for_us_gurps/
I switched to DFRPG in 2021, and it was a giant pain. You aren't supported in any way when creating encounters (the system doesn't try to predict encounter balance or player power), and you aren't supported in any way when trying to figure out how much wealth to dole out.
Compare this to PF2e encounter building or wealth by level.
The default magic system is also massively imbalanced from a game perspective, so the wizards can't expect to pick flavorful spells and be good; they have to actively avoid the duds.
Starting at 250 points also is about equivalent to level 6 D&D characters, which is disorienting. You could start them at 125 points (which is about level 1), but now you're off-template in the book.
It's hard to play the resource attrition game that 5e GMs are used to: rather than spell slots, casters have FP/Energy reserve and that comes back very quickly. Martials tend to not care about FP, so lots of the mechanics don't apply to them. They don't care about FP loss from not sleeping, FP loss from hiking, FP loss from weather, whereas casters care a lot.
The merchant buying/selling system is heinously convoluted: you get a different price for each item based on
The seller's wealth level
The buyer's reaction score
A haggling roll (on each item)
How much you identified the item to be worth (which uses 4 separate skills; merchant, Connoisseur(art), Connoisseur(luxuries), Connoisseur(weapons)). If you fail your identification roll, you think the item is worth a GM-specified price (which now you have to get familiar with). There are cases where failing is beneficial.
The encumbrance system is extremely unwieldy: tracking weight down to 0.0x pounds (each coin weighs 0.02 lbs), same with other gear.
The enemies dying system is also unwieldy: whenever they're below 0 HP, they make a HT roll to stay conscious. If they take at least half of their health in a single blow, they also make a roll to stay conscious. Players need to learn to target the skull/vitals/face to cause unconsciousness otherwise fights drag.
There's a massive amount of skill bloat; DFRPG includes stealth (being hidden), camouflage (blending in with the background), and shadowing (following someone undetected). Those can all be the same skill, and are in every other game I've seen. Players wanting to play stealthy characters suffer a skill tax as a result.
Mimicry (animal sounds), Mimicry (bird calls), and Mimicry (speech) are different skills. Merchant (haggling with merchants, identifying the value of gems), Connoisseur (art) (identifying the value of art), Connoisseur (luxuries) (identifying the value of cloth/perfume/wine/etc), and Connoisseur (Weapons) (identifying the value of weapons/armor) are different skills.
Boating and Seamanship are different skills. Running, Lifting, Swimming, Throwing, Jumping, and Climbing are different skills. Sleight of Hand, Smuggling, and Filch, and Holdout are all different skills for stealing things. Observation, Perception, Scrounging, and Search are all different skills for finding hidden things.
There's also next to no support for published adventures or adventure paths. Gaming Ballistic has the Nordlond stuff, but I consider that to be very far away from being ready to run at the table. None of what I've seen holds up to modern adventure design
https://beyondfomalhaut.blogspot.com/2016/08/review-standards.html
https://princeofnothingblogs.wordpress.com/reviewing-standards/
You can homebrew all of this stuff: fix selling things, change how travel works, change encumbrance to track item slots, merge skills together, invent wealth-by-level tables, and invent systems to give monsters/encounters difficulty ratings, but it's all you. I did all of these things; it was exhausting. I'm not allowed to publish (for free) what I did because of SJGames licensing. Frustrating.
On Game Design Influencing Negative Behavior in Cooperative Games
https://www.reddit.com/r/Games/comments/112ypgo/riot_forge_games_2023_the_year_ahead_trailer/
Imagine of version of league of legends where you win by either 1) destroying their nexus 2) getting the first kill. As soon as the first person on a team dies, the game is over and they "caused" the loss. You might have these 14 minute games where you're doing really well in your lane, building up power, having an econ lead, etc, and then Ahri dies mid and your team loses. I would imagine that game would be more toxic than regular league.
Or, imagine a version of league of legends where XP and wealth were socialized: everyone on your team is always the same level and has the same amount of wealth, so you can't have 1 character "get fed" or "fall behind".
Or, imagine a version of league of legends where getting player kills didn't reward money or XP, all farm came from mobs/towers. Before, someone making a bad decision and dying 1v1 top meant feeding top lane and falling behind, and losing map control. Now it doesn't.
Or, imagine a version of league of legends where there's no XP/econ system. You level up according to the game clock and your item build is just baked into your level. Now, some idiot doing dumb stuff in top lane doesn't affect bottom lane at all; you will always be equal level/power to your opponents.
All of these would still be competitive MOBAs! Yet, I’d expect for them to have different amounts of negativity-toward-teammates.
Summary of GURPS Dungeon Fantasy Issues
https://www.reddit.com/r/gurps/comments/11rbku6/gurps_dungeon_fantasy_in_play/
Howdy! I switched to gurps dungeon fantasy in 2021, and my last session with it will be in a few weeks. I started playing 3.5 as my first TTRPG back in 2004, then eventually moved to pathfinder 1e, then dnd5e, then pathfinder 2e, then gurps over the years. My next campaign is 13th age (I would personally have preferred to GM OSE, but my friends love tactical minis combat).
What I Wanted
A magic system where you weren't punished for using utility spells
Combat that didn't feel "solved"; there ideally are mechanically interesting choices
Combats with quicker resolution (we had just come out of a years long PF2e campaign, and the upper leveled enemies start having a lot of health)
Generally better mechanics for out-of-combat resolution
Better-supported attrition. In PF2e there isn't any meaningful attrition - there's infinite out-of-combat healing via recharging focus spells and the medicine skill, consumables are garbage, and the published adventure paths enable you to rest extremely frequently to recharge spells.
Notably I did not care about realism. I care way more about playing to genre tropes and especially fun-at-the-table. If something is "more realistic" at the cost of being less in-genre, or a slog to actually accomplish at the table with physical dice, pen, and paper, then I think it's not for me.
Did It Deliver?
Broadly, no.
A magic system where you weren't punished for using utility spells
This did work. In D&D, if you're playing a caster and want to cast comprehend languages in order to read a tome, you either need to
prepare that in advance (and thus you're weaker for effectively missing a spell slot)
wait until the next day, prepare it, and cast it (so when you're exploring you're fine, but now you're waiting and weaker tomorrow)
learn it as one of your limited spells as a spontaneous caster
My table ended up hating spells like comprehend language and it turned into "okay the GM gave us a book we can't read; i guess we spend money to buy a scroll". I wanted casters to feel magical and cool; spending FP (which quickly replenishes in gurps dungeon fantasy, more about this later) to cast solved the problem. Casters were free to cast stuff like Resist Disease to deal with awful sewers, levitate to deal with big pits, etc, and not feel like they were giving up combat viability.
The downside is that there really quickly became no way to have caster attrition. After a few sessions of play, the wizard in the party had 11 FP, 10 ER, and Recover Energy 20, so he had effectively 21 mana, got a 2-mana discount on all his spells, (which made maintaining a lot of them free), and recovered 1 FP and 1 ER every 2 minutes (10 mana per 10 minutes, effectively). He could lend energy to anyone else and then recover ER while marching. This means that the typical genre tropes of folks getting tired and worn out were gone.
Combat that didn't feel "solved"; there ideally are mechanically interesting choices
I thought that with the hit locations, deceptive, rapid attack, retreat, all-out attack, shock, stunned, etc conditions, that combat would be interesting and people would have to keep adapting to new situations that encourage improvisation and dynamic play. That isn't how it worked out for my table.
Instead, we had a Barbarian with 22 Axe/Mace and a Scout with 26 effective Bow skill. The wizard could maintain Haste-1 on the party for free, and would crank that up to Haste-3 (+3 dodge) and the cleric would drop Shield-5 on the Barb and they could both just dip. The Barb would have 18 dodge (1% chance to get hit), luck (to avoid getting critted), ~8 DR (so most attacks wouldn't hurt him from getting ganged up on), and the cleric could bless in advance.
The Barb could just swing with ~4 ranks of deceptive with his maul for 3d+8 damage, and most of the stuff in the bestiary couldn't defend in any effective way. That would bring it down to half move half dodge, and then he could target the skull and it would die. The archer could shoot eyeballs from 10 yards at a 16 twice a turn. Dodge or die.
Combat got solved very quickly and in a degenerate way: they stopped taking damage and using resources.
Combats with quicker resolution (we had just come out of a years long PF2e campaign, and the upper leveled enemies start having a lot of health)
But it still took a while. Every attack needs to be defended against. Facing matters. The defender can retreat. Adding up 3d6 takes longer than looking at a d20. You have to declare your hit location and ranks of deceptive. There are a bunch of numbers to keep track of (did this guy block aready? have they parried? what's their posture? what's the light level? am I shooting from 30 degrees or 45 degrees? range penalty for 5 yards is different than 7 yards). Roll a random hit location. Roll random crit effects. Once they do damage, you have to subtract DR. Then, multiply by wounding modifier which varies by 5 different types and also varies by location (cutting does 1.5x damage to the torso but 2x damage to the neck, piercing does 0.5x damage to the torso but 3x damage to the vitals). Roll for knockdown/stun (modified by location). Roll to stay conscious when crossing HP thresholds. Roll to stay conscious for each enemy at <= 0 every turn. Don't forget traits like No Vitals, Homogenous, or Diffuse. Don't forget traits like Fit when rolling for consciousness. AAAAHHHHH
Generally better mechanics for out-of-combat resolution
My thinking has evolved a lot here. I wrote up Proficiency Splitting a while back and stand by it; the more you split up characters (splitting up "thievery" into "filch" - stealing medium-sized objects and "pick-pocket" - stealing small objects), the more you can have mechanically differentiated character sheets, but now you have to worry about more mechanics and the more points you have to invest to be broadly good at stealing things (oh sorry, you're a buffoon when it comes to swiping coins off a desk - you took filch instead of pick pocket). Is it more realistic? Sure. I'm skilled at pen-spinning and some sleights of hand in real life, but bad at coin magic. Doesn't fit the genre though IMO.
These days, I prefer really simple out-of-combat resolution mechanics. Advantage and Impact - Dreaming Dragonslayer is an awesome thought piece. I'm fine just saying "okay you want to try to swing across this gap? sure, lets roll a d6 and on a 3 or higher you succeed. Otherwise the rope will snap and you'll have a chance to grab the far ledge." and then go from there. It doesn't feel like there's any more of a "game" to it when you add in a bunch of modifiers - at the end of the day it's just a probability. You're going to roll the dice and get a result; I just want to make sure that the folks at the table think the probability is reasonable and we can more forward. Hundreds of pages of gurps crunch is one way to accomplish that, just making it up and then talking about it is another.
Better-supported attrition. In PF2e there isn't any meaningful attrition - there's infinite out-of-combat healing via recharging focus spells and the medicine skill, consumables are garbage, and the published adventure paths enable you to rest extremely frequently to recharge spells.
On the surface, the gurps attrition looked good: you'd have slowly recharging FP and each time someone got healed they would keep stacking up -3 penalties to being healed again. On a deeper look, it's a paper tiger. Characters are almost never getting hit. FP recovery becomes very fast. There are ~4 different healing abilities that don't share the same -3 penalty (so you can rotate them). Scrolls of Major Healing cost $160 and heal for 8 HP (which is almost all of a character's health). $160 is inexpensive; a +1 sword would run you $10k (but there's no guidelines at all for giving out wealth). Mirror of the Fire Demon (the first party starter adventure) gives out around $30k in post-sale proceeds (so enough for 187 scrolls).
The biggest thing though is how unsupported I felt as a GM. I talked about this here.
You aren't given guidance about how many monsters might show up in an area (so it's difficult to properly build encounters for the world). How many orcs are together? How many bandits in a bandit lair? How many giant spiders? You also aren't given guidance about how to balance encounters for a party. Say you want to create an even match between the players and some hobgoblins - how many do you need? PF2e and 13th age are really good about party-based encounter balance. OSE is really good about naturalism]. Roving dungeon bands of skeletons are in packs of 3d4, and are in dungeon lairs of 3d10. Roving wilderness bands of skeletons are in packs of 3d10 and their lairs have 3d10 x 5 skeletons.
You aren't given guidance about how much treasure you ought to be giving the PCs or how much treasure monsters often have. PF2e has stuff like this. There are similar tables in 3.5 and 5e.
There isn't an active gurps module-writing community. 5e has the DMs Guild, which has amazing stuff like what the Arcane Library is putting out. The OSR community has a boatload of reviewed adventures held up to very high standards. Gurps has the gaming ballistic stuff, which I find to be difficult to run.
On How Not All Fighting Game Matchups are 50:50
https://www.reddit.com/r/SSBM/comments/14sfnvn/kodorins_midlevel_marth_tier_list/
Matchups do not really matter in this game. My opinion. All that matters is who is playing their character good enough to beat their opponent.
This feels like a super hot take!
If you're not familiar with the field, check out mathematical game theory.
There's this situation that happens in soccer where one player will have a penalty kick. The way it plays out is that pros will either aim for top-left, bottom-left, top-right, or bottom right. Their kicks are fast enough that goalies can't wait to react to go left or right, so they have to commit. Even if they guess left/right fast enough, they can still miss top/bottom if the kick is good enough.
To make it more complicated, right-footed kickers are normally better at kicking left than right.
Here's roughly what it looks like:
| Kick/Goalie Dive | Left | Right |
| ---------------- | ---- | ----- |
| Left | 50% | 100% |
| Right | 100% | 20% |If the kicker kicks left and the goalie dives left, it might go in 50% of the time, whereas if you kick right and the goalie guesses right, it only goes in 20%.
The big question is: "how often should the kicker kick left and how often should the goalie dive left?"
Say the kicker always kicks left. It's optimal for the goalie to always dive left. Around 50% of kicks will go in.
Say the kicker kicks left 50% of the time. Say the goalie always dives left. That would be .5•5 + .5•1 = 0.75, so the kicker scores 75% of the time. Say the goalie also dives left 50% of the time. That would be
.25•.5 + .25•1 + .25•1 + .25•.2 = 0.675, so the goalie gets a little bit of value by mixing up left/right.
What's optimal for the goalie? Always diving right. Makes it so the goalie is able to save 40% of the goals.
What's optimal for the kicker? We want to pick a go-left percentage where the goalie's best distribution is the most favorable one for us. If we always kick left, the best the goalie can do is always jump left and so we score 50% of our goals. If we kick left 50% of the time, the best the goalie can do is always jump right and we score 60% of our goals.
The best the kicker can do is to kick left ~62% of the time. If the goalie always dives left, they score 0.62•.5 + .38 = 0.69; 69% of the time. If the goalie always dives right, they score 0.62 + .38•.2 = .69; 69% of the time. If the goalie splits 50/50, they score 0.62•.5•.5 + .5 + .38•.5•.2 = 0.69; 69% of the time.
We call that the mixed equilibrium - the goalie's choice doesn't matter any more since the kicker is scoring 69% of the time no matter what! That's what I mean by playing close to game-theory-optimal - if the kicker is kicking left anything but 62% of the time, they're losing value and could be handling that micro-situation better.
Way back up in melee-land this is relevant to matchups because a tournament set is a sequence of micro-situations. There are things we can react to and get value out of, and then there's unreactable mixups (especially in neutral, or DI mixups during punish).
Better characters are going to have more options (make characters guess three or four ways instead of just 2), and those options are going to be more rewarding if the opponent guesses right less rewarding for the opponent if they guess wrong. Better characters make you guess more often.
Doing analysis on these micro situations (what options does player 1 have? what options does player 2 have? how do these options interact in terms of likelihood to win the game from there? what's the game-theory-optimal way to play this situation? what situation does it lead to?) creates a markov chain.
Translated, discovering matchup technology always involves either 1) discovering a new option in a micro situation or 2) finding a way to turn something that was previously a guessing game into a reaction.
We don't normally use these terms (and for good reason; this is pretty esoteric stuff), but you'll hear top players talk about matchups in terms of "tools" or punish and neutral, etc. Those are ways to categorize the above math - when a character has better neutral, that means they're either more effective at putting the game into a favorable micro-situation, or that they're capable of reacting when other characters would have to guess. When a character has better punish, it means that their reactions are more rewarding or their micro-situations are more rewarding when they guess correctly.
I totally get that the analysis is hard! But I really don't get the take "every matchup is even or 100-0". There's a whole spectrum in between!
On The Weirdness of Quest-Driven Crafting
https://www.reddit.com/r/rpg/comments/16anl87/what_you_likedislike_in_ttrpg/
Player: I want to craft a +2 flaming sword
GM: Okay, you'll need to find someone that can tell you how to do that.
Player: I try to find someone that can tell me how to do that.
GM: (scrambling) You learn that Huxley the Heralded might be able to tell you how to craft such a sword.
Player: Okay, I go to talk to Huxley the Heralded.
GM: You don't know where Huxley the Heralded is, but perhaps you can find someone that knows where to find them.
Player: Okay, I look for someone who can point me to Huxley the Heralded.
...
(many weird improvised subquests later)
GM: Huxley the Heralded tells you that in order to craft a +2 Flaming Sword you need bathe the sword in the blood of a titan heart and quench it in an active volcano.
Player: Okay, where is a titan? Some time later: Okay where is an active volcano?
It starts to feel real fetch-questy and contrived. Coming up with good adventures on the spot is legitimately tough (which is why 90% of published adventures are garbage). Making it so that you have to improvise a whole adventure every time the player wants to craft anything seems like a really good way to give yourself a ton of work or tell players to not craft, in my experience.
Chris McDowall writes in Trade Without Money that in mythic bastionland:
Only the rich deal in coins. Most trade relies on more practical matters:
• Finding somebody who can supply
• Securing raw materials if needed
• Giving something in return, or owing
The value of any given item depends on the bargaining positions of the person holding it, and the person desiring it. Long-standing trade agreements often form into pledges of ongoing service or protection.
This seems like a really fast way to make very boring gameplay (though, I could be thinking about it all wrong).
Player: I'd like to acquire a bow from the bowyer.
GM: The bowyer is all out of bows, but they could make one for you if you provide sinew and yew wood
Player: I go acquire sinew and yew wood
At this point, the GM either handwaves it so that actual gameplay can continue, or you roleplay out talking to the lumberyard for yew (or finding a yew tree and cutting it down), and talking to the butcher for sinew (and figuring out what fetch quest they'll send you on) or hunting an elk yourself or whatever. It doesn't seem more fun to me than just buying a bow, or being able to craft a bow with "~x gold worth of materials and y amount of downtime".
On the Hidden Woes of Rule Ambiguities
[Knave is] never pretending to be a robust, crunchy system for miniature chess, yet some of your arguments feel like that's what you're expecting, and that comes off as weird and potentially disingenuous at times... The fixation on defining melee. The word is used in its literal meaning, it defines itself. If you've ever read any fantasy literature or played any kind of fantasy adventuring game, you know what melee means.
Yeah, totally get that melee combat is the sort of combat where we're attacking each other with melee weapons. In FATE, you just need to be in the same zone.
In wargaming, minis have movement speed measured in inches, are considered "in melee" when their bases touch. In 13th age, the GM defines "close, medium, far", and you can move between distances with a move action, weapons have the same ranges. 3.5e explicitly goes over the 1" grid, and mentions that melee attacks can hit adjacent squares, and provides a picture of this. GURPS uses 3ft hexes, and mentions that melee attacks can hit adjacent hexes. In the PBtA games, you just narrate yourself moving into melee combat, and the ref decides whether or not that triggers a move (like defy danger).
Knave 2e defines move speed (unlike fate, 13th age, ptba), but like wargames, 3.5e pathfinder, and gurps. It also states that monsters start 2d6•10 feet away, so we have numerical specifics. It never mentions a grid. Say the enemy is current 44 feet away, and your character can move 40. Can you attack? Yes in 3.5e, no in gurps. Say that your character is 6 feet away, are you "in melee" for the purposes of being able to shoot a bow? No in wargaming, gurps, and 3.5e, yes in fate, 13th age, and ptba.
We know Knave is a BX derivative, and inherits more from the OSR than it does gurps, so I'd likely make the common law that we're using 1" squares and that melee is being in an adjacent square (or <=5 feet away), but I think it would have been better to define it.
Assuming that the GM is unable to fairly rule what counts as a sneak attack.
I find this is a very difficult, contentious topic. If me and a buddy have a full flank on someone (one stands directly in front, one directly behind), do either/both/neither of us get to sneak attack? In GURPS, minis having facings, and a character out of your line-of-sight can sneak attack you unless you have special senses, over and over in the same combat. Can someone sneak attack, then hide again next turn, and sneak attack the same victim in the same battle? PF2e has rules for this. How much cover is required to be able to sneak? etc.
I played a DCC valentines one shot a few weeks ago (as a player), and the very first thing that came up was the thief player discussing with the GM whether or not he could use his action to try to hide to be able to backstab again. Backstab did so much more damage for him (just like in Knave), that it was worth it for him to pause the game mid-combat and hash out how this worked. If he got an unfavorable ruling, he would have likely chosen a different character.
This stuff matters!
The fixation on players somehow requiring certainty in how long exactly they can explore with regards to torches and/or rations. It's an intentional design decision that it's unpredictable, and candles and foraging mitigate that unpredictability, which I think deserves a specific mention if you think the unpredictability is a problem. (Relatedly I have to add, I share many of your qualms with hazard dice, well articulated points with that.)
I don't think it's fixation; rather just analysis. You can forage (with roughly the same success rate in OSE), so instead rations are higher variance in Knave compared to BX. Given that the upside of the variance is low-reward and the downside of the variance is high-risk, I posit that the players will learn to hedge against variance by carrying a lot more rations and torches than normal, and then calculate just how costly that hedging is (I'd guess that the author did not perform this statistical analysis).
The implied point I'm making is that adding variance has player behavior consequences, and those consequences are worth analyzing when you're analyzing the mechanic. Is it actually making the game better to have rations last a highly variable amount of time? For what it's worth, my home game doesn't distinguish between fresh and iron rations at all because I go in the total opposite direction. I want folks spending their mental energy on other, more fun things than thinking about ration expiration times or variance.
The interpretation of the overland brigand encounter feels deliberately obtuse
"Day 1 Watch 3: Travel to D7. Roll 1d6 (1). Encounter. Roll 4d6•30 (450ft) for distance, 2d6 (5; Insult, threaten, or command the PCs), 1d12 for type (3; 1d8 brigands), and amount 1d8 (7). Have a grand ol’ fight against 7 Brigands that are insulting, threatening, or commanding the PCs from 450ft away."
I needed to roll all the dice to be able to sum up exactly how many dice had to be rolled in order to make the journey. This is actually the same number of dice as in BX (which I think is also a problem, and is a spot that a potential new game can streamline). The major difference is that in BX, the bandit's reaction would be more vague (and easier to apply generally). A 5 is "hostile; may attack", which removes a lot of my ability to be snarky. Notice how "Have a grand ol' fight against 7 Brigands that are hostile, and may attack the PCs from 450ft away." makes way more sense as a sentence.
In Knave being more specific, it creates more opportunity for the result to make less sense (like insulting, threatening or commanding from long distance). In being less specific, the BX table is easier to figure out how to apply to the fictional situation.
Insistence on explicit definitions with combat actions, like whether holding actions etc. is possible.
These come up in every single game I play! If an archer can move out of cover, shoot, and back into cover, that makes positioning archers at walls massively better than if they can't. In BX, you have to move before you shoot, and so we know archers can't do this.
If melee combatants can move, attack, and move, then you can have a shield line with 2-handers standing behind it. The high-dps characters can run out, attack the enemy, and run back to safety behind the shield line every turn.
If you're trying to hold a cooridor against a zombie horde, the zombies can attack and then move out of the way for the next zombie, so the front-liner is potentially taking an attack from all ~30 zombies instead of just the front 2 (aka the zombie shuffle at my table).
This stuff matters!
Carousing mechanics being such a big deal. Surely the referee has the power to control pacing and the player has no authority to force a full-blown, detailed prison-break scene to "derail" our ability to go run the module. One roll whether they succeed in escaping, if they fail, they pay the bail. You're done in 2 minutes. If they didn't have the money, they're in debt now, better go find loot.
The player certainly can't "force" anything. What they can do is request to go carousing, and then the dice result in someone being imprisoned. The GM (and table) can choose to ignore the dice and say that didn't happen, but the same can be said of any dice roll (attacks, damage, saving throws, etc). Once the players are in prison, the GM can handwave the whole ordeal if they like "yeah, you were imprisoned, but you paid, say, 50c and now you're out; what do you do next", but why are we doing this then?
Insistence on shopping not being completely eliminated because PCs can take money at start if they want to.
Both the original post and this one have mentioned me of being stuck on shopping not being totally gone. That isn't my hangup (though I totally understand how one could read it that way). Rather, I'm responding to the developer commentary that "Careers are a quick way of giving players a background and a starting loadout without having to do a lot of shopping."
As far as I can tell, they aren't! Reviewing the career starting items, they're not the sort of items I would choose to fill my precious slots with, and we're given money to choose different ones. I'd hazard that players will start choosing to ignore their career items and start buying more torches, rations, more useful tools, etc.
As in, if the goal was to make this quick and reduce shopping at chargen, the only change we'd have to make here would be to take the money away. Then players really are doing less shopping, and actually using the tools they randomly generate rather than having to make complex logistical choices.
On Clarifying BX Reaction Rolls
https://www.reddit.com/r/osr/comments/1duhqqb/please_help_me_learn_to_run_ose_better/
I didn't know how to handle reaction rolls that weren't hostile.... I feel like my party and the monsters are just staring at each other waiting on me to make a decision. I don't know what to do to resolve an "Uncertain" outcome.
I found the verbiage in the text to be very unhelpful. A lot of ODND and BX are unclear or left to the GM to figure out how they want to it to work. As an simple example of this, how does "I want to trip the hobgoblin" work?
Here's what page 24 of Basic says:
PARTY ACTIONS: The first decision a party must make in an encounter is whether to fight, talk, run or wait to see what the monsters will do.
If the party chooses to fight, combat will begin [...]
If the party chooses to talk (and the monsters will listen), the DM plays the part of the monster. The players can ask questions, make bold statements, and otherwise react to the creature. The encounter may then become peaceful (agreement!) hurried (as the monster or party runs away), or violent (if the talk leads to combat).
If the party chooses to run away, the monster might not follow, in which case the encounter is over. [...]
If the party chooses to wait to see what the monster will do, the DM must decide the monster's reactions
Then, we have our famous chart:
| 2d6 | Reaction |
| ---- | --------------------------------------------- |
| 2 | Immediate Attack |
| 3-5 | Hostile, possible attack |
| 6-8 | Uncertain, monster confused |
| 9-11 | No attack, monster leaves or considers offers |
| 12 | Enthusiastic friendship |(to help with memorizing the table, the sequence is 1 result, 3, 3, 3 results, then 1 result again. Symmetric. So you can count 2: worst. 3,4,5: bad. 6,7,8 medium. 9,10,11 good. 12 best.)
So, as far as I can tell, the players make a choice from 4 options: fight, talk, run, wait. Monsters either want to fight, talk, run, or ignore the PCs.
Here's how I'd represent this:
| 2d6 / Action | Fight | Talk | Run | Wait |
| ------------ | ----- | ------------- | ------ | ------------- |
| 2 | Fight | Fight | Chase | Fight |
| 3-5 | Fight | Fight if adv. | Chase | Fight if adv. |
| 6-8 | Fight | Talk | Ignore | Ignore |
| 9-11 | Run | Talk | Ignore | Ignore |
| 12 | Run | Talk | Ignore | Talk |So if the players try to fight the monsters, the monsters fight back on a 8 or less, or run on a 9+. If the players want to talk, the monsters will on a 6+. If the players want to run, the monsters let them go on a 6+.
For talking, after each exchange, you can reroll entirely or interpret the reaction roll as shifts in their disposition. So a 2 means 2 shifts toward the immediate attack result, a 3-5 means 1 shift toward immediate attack, 6-8 means 1 shift toward uncertain, 9-11 means 1 shift toward friendship, and 12 means 2 shifts toward friendship.
I like the shifts because that way their behavior is a little smoother. You don't go from hostile to enthusiastic friendship with one roll.
On Rules Light Games
https://www.reddit.com/r/osr/comments/1dy7sax/is_a_lightweight_ruleset_a_good_choice_for_an/
All sorts of D&D-like games need to be able to answer all sorts of questions:
Can I leap this 20ft gap?
Can I sneak past this guard?
Can I bribe this guard to let me past?
Can your party defeat these monsters?
Can I scoop up some sand and throw it in their eyes? What does that do?
Can I shoot from behind cover?
I suspect there might be something hidden in the desk - can I search it?
The chest is locked, can I pick it?
The door is stuck, can I force it open?
Can I run away from this fight?
Can I climb this wall?
How many Orcs do we encounter?
What treasure does an Orc lair have?
Do we ever grow more powerful? When? In what way? I'd like to be able to fight Dragons and Giants one day!
These are all common questions that come up over and over in D&D-style adventures. Broadly, there are two ways to answer such questions.
1) The system can create explicit rules that help cover as many of the questions that are likely to come up in play as possible. It can further try to write the rules with more and more clarity. Then, the task of answering a question that comes up in play is boiled down to looking up the appropriate rule. Everyone at the table can be shown the rules beforehand, and everyone can help look them up or know the system. The more questions a system tries to answer, the more unwieldy it becomes. It's harder to find the answer to your question in a book of 1000 answers than 10 answers.
2) The system can create broad frameworks and attempt to express it's intent to provide guidance for you to come up with your own answers to such questions. Once you've done this, it becomes precedent going forward. Typically you do so from a combination of the game's guidance and your (and everyone else at your table's) invisible rulebooks.
No game can attempt to cover all questions and scenarios explicitly, so eventually the GM has to make a ruling. The less rules it attempts to cover and the more broad-strokes it uses, the more "rules-lite" it is and vice versa.
As a demonstration of this, compare a Thief hiding in OSE to one hiding in Into the Odd. In OSE, it gives you a specific percentile chance based on your level (so a 6th level thief has a 36% chance to hide in shadows). In Into the Odd, you might just make a DEX save just like any other vaguely dexterous thing; there aren't specific hiding rules.
The main things I think rules-lite games struggle with in the context of long-form games are
How do the upper levels of play work? Can the game run Against the Giants?
How do higher level spells work? Do the players have answers to high-level ailments like petrification, curses, being polymorphed, death, etc? Are they able to teleport, scry, ask questions of gods, etc?
How does campaign play work? How quickly and far can I move my army? If my army fights another army, what happens? How does sieging a castle work? How much does my army cost to maintain? Are my peasants part of the game, and if so, how?
So! A lot of folks are equipped with invisible rulebooks that help make rulings for low level play. Maybe you have a good handle on how far someone can jump, and so a ruling you make will be fair. If you don't, and your rulebook doesn't cover it, maybe you import some rules from another system that seems to jive.
Less folks, I think, are equipped with invisible rulebooks that give them a good handle on high-level play. It's very unlike regular human experience (whereas low-level play is relatable). So, if your system doesn't cover such questions, it may be helpful to import some structure from ones that do.
On Tactical Combat in OSR Games
https://www.reddit.com/r/osr/comments/1e5yhy6/osr_game_with_the_best_tactical_combat/
I have gone on record in the past saying that I think most OSR games discourage combat, not because its dangerous but because its boring.
I've thought about this problem a lot, but don't normally see it get talked about.
For instance, we have Information, Choice, Impact - Bastionland that tells us that gameplay is about making impactful choices with sufficient information.
So why is it that in combat, there's no information about how maneuvers work (knocking someone down, shoving them back, running past them, how thief backstabs work, throwing sand in the eyes, disarming, etc) and of the choices the players do have, there's relatively little impact (a player skilled at BX combat is very similarly effective as a new player)?
You're going to get (and have gotten already) responses saying that the tactics and the choice come before and after combat - when to fight, how to parley, etc. That's hogwash. Those also are choices but it's not what you're talking about.
In terms of pre-existing OSR games, I don't think you're going to get there. If you want pre-made tactics games, then 13th age, lancer, D&D 4e, or pathfinder 2e is the spot. It's probably easier to rip out those games' non-combat systems and give them OSR structure than it is to replace OSR combat with tactical combat. I did it with 13th age successfully.
But! If you do want to keep OSR games and introduce some tactical elements, here's what I suggest:
Cleaves (additional attacks after downing a foe). Characters using the fighter thaco progression get 1 cleave per class level, and characters using the thief thaco progression get half a cleave rounded down per class level. A character can take a 5ft step before each cleave but can't move more than their regular movement each round. Works for melee weapons and missile weapons in short range.
Damage bonus. Characters using the fighter thaco progression get +1 damage at level 1, and an additional +1 damage every time their thaco improves. Thieves get an additional dice of backstab damage at level 1, and an another additional dice every time their thaco improves. This really helps with making Fighters feel quadratic and not useless next to Wizards at higher levels. Makes it so that fighters and thieves can cleave through groups of low level mooks.
Charges. Characters can move 3x their speed in a straight line and attack with +2 to hit, but have -2 AC until their next turn. A charge with a spear-like weapon rolls an extra dice of damage.
Delays. Any time during a character's turn, they can delay the rest of their turn (so either their attack or their movement) until they say so. This allows them to take the rest of their turn after an enemy moves but before it attacks, for example. Handles stuff like readied attacks, moving in place then waiting to be passed an item and then using it, setting up flanking, etc. A delayed attack with a spear-like weapon against a Charge rolls an extra dice of damage.
Facing. Assume that a character is "facing" whatever it's attacking. They're Vulnerable to attacks from behind it (-2 AC and enables Thieves to backstab). When an unengaged target becomes engaged, it can face its attacker for free.
Threatened Squares. A character with a held melee weapon threatens its adjacent squares. If any enemy of that character enters any of those squares, it must stop movement and becomes engaged. This makes positioning and terrain much more important.
Firing into melee. No shooting at engaged targets, flat out. Shooting past an engaged ally to hit an unengaged foe is at a -4 penalty. Get into melee!
Maneuvers. Roll at -4 to hit, and if successful do your normal damage. Then the target makes a Save vs Paralysis and if it fails the thing you want happens. This handles shoving, pulling, knocking prone, blinding for a round, running past threatened squares, etc. This is usually what zombies should be doing when they're stuck in a choke.
Prone. A prone creature is Vulnerable. A prone creature can attack at a -4 penalty. If they're engaged, they can stand in lieu of attack. If they're not engaged they can stand in lieu of movement.
Vulnerable. Thieves can use their Backstab against Vulnerable targets. A vulnerable creature has -2 AC (does not stack with the +4 from Backstab).
Morale. The current BX system has you rolling twice, and so the chance that everything fights to the death is still unreasonably high. I suggest scrapping it and just almost always having enemies flee before they die. An individual probably shouldn't stick around once it has lost 2/3rds of it's health and a group probably flees after losing 1/3rds of it's members. Their goal is to get friends and come back with at least 2:1 or 3:1 odds. A good example of this is watching how professional SC2 players micro their units. They will make low-health units retreat and heal up at home to fight another day.
The other bit is that you can reintroduce a lot of the combat resource management game that happens in lancer/4e/pf2e/13th age with custom magic items.
For instance, instead of a +1 sword, you can have a sword that gives +3 to hit and damage when they speak the command word, and they can do this after rolling (to adjust the result). Works once a day (or once per 10 minutes for a stronger version).
A mace where successive hits charge up the mace, increasing it's damage by 1 each time for the rest of the fight.
IE, instead of flat numerical bonuses, create magic items that act as per-fight or per-day resources. Can pull from tactical games (even stuff like final fantasy tactics), MMORPGs, Gloomhaven, etc for good examples.
On the Risk:Reward of Wilderness Lairs vs Dungeons in BX
https://www.reddit.com/r/osr/comments/1ei7gr3/bx_wilderness_is_more_profitable/
A couple of notes:
Monster listings note which treasure type is present in the creature’s lair and which it may carry on its person. See Treasure Types for details on how to roll treasure according to a treasure type letter code.
So, treasure is found in lairs unless they specify personal treasure like how a Bandit has type U treasure when found in a random encounter and type A treasure in its hoard.
Then we have Random Room Stocking
Treasure: If a monster is in the room, roll the treasure type indicated in its description. Otherwise, the treasure depends on the dungeon level (see below).
Treasure is either guarded by a monster, in which case you use the monster's treasure type, or unguarded in which case you roll on the treasure in empty or trapped rooms by level table
From this we can infer that a stocking result of monster+treasure refers to a dungeon lair, and they have their lair treasure. This is further supported by the description of Treasure Type
A to O: Indicate a hoard: the sum wealth of a large monster or a community of smaller monsters, usually hidden in the lair. For monsters with a lair encounter size (see #Number Appearing) of greater than 1d4, the amount of treasure in the hoard may be reduced, if the number of monsters is below average.
Finally, from Number Appearing:
Listed as two values, the second in parentheses.
Zeros: If the first value is a zero, monsters of this type are not usually encountered in dungeons. If the second value is a zero, monsters of this type are not usually encountered in the wilderness and do not usually have lairs.
Usage: The use of these values depends on the situation in which monsters are encountered:
Wandering monsters in a dungeon: The first value determines the number of monsters encountered roaming in a dungeon level equal to their HD. If the monster is encountered on a level greater than its HD, the number appearing may be increased; if encountered on a level less than its HD, the number appearing should be reduced.
Monster lair in a dungeon: The second value lists the number of monsters found in a lair in a dungeon.
Wandering monsters in the wilderness: The second value indicates the number of monsters encountered roaming in the wilderness.
Monster lair in the wilderness: The second value multiplied by 5 indicates the number of monsters found in a lair in the wilderness.
So there are 5 different types of monster encounters.
Random encounters in a dungeon (the first number, no treasure).
Stocked non-lairs in a dungeon (the first number, no treasure).
Stocked lairs in a dungeon (the second number, with always treasure).
Random encounters in the wilderness (the second number, no treasure)
Monster lairs in the wilderness (never random, the second number times 5, always with treasure).
Putting this all together, there are three different ways to find treasure.
Unguarded in a dungeon (roll based on dungeon level)
Guarded in a dungeon lair (treasure based on monster treasure type, guarded by the second number of monsters)
Gaurded in a wilderness lair (treasure based on monster treasure type, guarded by the second number of monsters times 5)
In practical terms, a dungeon might have orc random encounters (with 2d4 orcs), and might also contain an orc lair within the dungeon. Such a lair has 1d6•10 orcs (~35 orcs) and has on average 3900g. A random orc encounter (of that same 1d6•10 orcs) in the wilderness is valueless. A orc lair in the wilderness has ~175 (1d6•50) orcs but the same 3900g of treasure.
Thus! Loot in a dungeon is less heavily guarded, encouraging players to seek dungeons rather than lairs.
BX Combat Analysis and Q&A
https://www.reddit.com/r/osr/comments/1elyr1s/my_questions_with_the_bx_combat_sequence/
So are we supposed to interpret them as having started these declared actions at the start of the round even if they lost initiative? If so, why isn't it only the retreat - the melee movement that causes the opponent to have a bonus to their attacks - that needs to be declared? What is the purpose of declaring the fighting withdrawal?
There are two sorts of melee movement - retreats and fighting withdrawals.
You have to declare retreats because combatants have a +2 to hit against you this round. Say that you didn't have to declare retreats before rolling. If you lose initiative, then the monsters attack you and the GM wouldn't know that they were supposed to have a +2 to hit. If you win initiative, it didn't matter.
You have to declare fighting withdrawals because it commits you to not moving forward. If you lose initiative and the monsters do something that would make you want to press on (like start running away), you cannot.
What is gained by declaring [a fighting withdraw]?
If you start your turn in melee with an opponent, the only way to move is to fighting withdraw or retreat. Retreats let you go further, but give attackers +2 to hit vs you. So, a withdraw is slower but safer. Additionally, retreat specifies "the retreating combatant may not attack", but withdraw does not have any such clause. Therefore, if after retreating you're capable of making an attack (like you retreated into melee range of another foe, or you have a ranged weapon), you can still attack.
My understanding is that these actions - spellcasting and melee movement - are meant to be tricky to carry out and more complicated to pull off unless you win initiative - and to incorporate that risk in the rules, must be declared before initiative is rolled. Is this a correct assessment - is this the benefit of this design?
My read is that you have to declare spellcasting to provide counter-play against spells. If the caster loses initiative, they can be attacked (like by a thrown dagger, bow shot, or meleed if possible) to disrupt them.
For melee movement, my read is that it's to make combat sticky. Since you have to declare melee movement, and the only sorts of movement you can do are backwards, it makes it much harder to run past the enemy lines to get to the archers and casters. Further, it prevents archers from effectively backpedaling to kill melee. Say that an archer is 30ft away from a swordsman. The swordsman wins initiative and runs up to melee the archer. On the archer's turn, they cannot move. They didn't declare melee movement, and could not have because you can only declare melee movement when you're currently in melee. It's weird, but this is how it works.
Side initiative seems to be entirely up to chance. It doesn't seem like there's anything that can be done to influence this die roll. A major factor determining how the fight will go is up to pure chance each round. This makes low level fights very "swingy" I would assume. Low-level unpredictability and deadliness of fights is accetable for me - is my assessment correct - is this the design intention?
I think swingyness is one of the intents! The more swingy combat is, the more low-level parties are able to kill dragons (and the more high level parties have to fear random encounters).
Side-based initiative also massively speeds up combat, and lets the players jointly come up with a coherent plan. Rather than having to do awkward turn-by-turn retreats, they all retreat as a unit. Rather than having to set up "i hold my hand out for bob to hand me a potion, and then delay my turn", "on my turn i hand alice a potion", "i trigger my delayed turn to drink the potion" shenanigans, it just happens.
I understand that all characters with slow weapons from all sides act last in the round, after every side's go has passed. Why is this not step 6 in the procedure? Are all slow weapon attacks simultaneous with each other by default?
Correct, and it probably should be. The rules are totally unclear what happens when there are slow weapons on both sides, but I think the most reasonable interpretation would be to add 2 more phases "winning initiative slow" and "losing initiative slow" after "losing initiative regular". Makes the numbered list less pretty though! That's how I'd code it if i was writing BX software.
someone pointed out that a lot of time is saved if it is rolled at the top of the round. Initially this seemed true as long as there are only two sides. If there are more than two sides, it still seemed to make sense to roll it before initiative. But after thinking about this further, if the PCs win the initiative, and significantly affect the lay of the battlefield, this may affect morale rolls, no? In that case, would it not be mechanically different to roll morale before the PCs act?
Rolling morale at the beginning of a side's turn (rather than before declarations) means that when the monsters fail their morale check, the ones in melee cannot flee. For example, the monsters declare no melee movement. The PCs win initiative and slay enough monsters to trigger morale. The monsters take their turn, fail their morale check, but cannot flee (except the ones that are not currently in melee). But yes, to your point, rolling it at the beginning of the turn rather than before initiative means that if the monsters lose initiative, their morale roll can be impacted by the PCs turn, whereas if you roll it before initiative it cannot be impacted (since the PCs turn hasn't happened yet).
If you're in a melee, you can only withdraw or flee, you cannot reposition within the melee, or otherwise move to join another melee.
The OSE rules clarify that you CAN move and attack. But if you're in a melee, you cannot move after having attacked, even if the result of that attack is that you are no longer in melee, as movement is always before attacks.
If you're not in melee, your movement is rather free.
I also noticed the absence of reach weapons in B/X (there's no second line with spears or polearms). So the only way to make non-missile, non-spell attacks is to be tangled in melee, severely limiting movement.
Am I summarizing this well?
All correct. It also means that archers cannot move after they shoot, so they cannot shoot and then move back behind cover. They're stuck where they shot from.
Another quirk is that the whole side's performs movement, then they all do their missiles, and then all of their spells, and then all of their melee. This means that the melee combatants have to commit to a spot before the arrows and spells go off. It's likely that they move up to a foe, only for the archers or spells to kill the foe, leaving them with nothing to attack when the melee phase comes around. Or, even more common, all of the melee combatants have to move together, and then attack together. If there are two goblins that are 20ft apart from each other, then Alice and Bob can either decide to both move to the same goblin, or move to different goblins. Bob can't wait to see if Alice kills her goblin to decide to move and attack the other one (unlike in individual initiative).
Notably, if Alice has a Slow weapon, she does wait to see if Bob kills their goblin before deciding where to move.
Some characters cannot move : those casting a spell cannot move. Those who are in melee and who did not declare they were fleeing or withdrawing cannot move intentionally.
Correct.
It seems that undead who were turned before they act must flee melee immediately during their movement phase, even if this was not their declaration. I suppose that others similarly affected by other spell effects and special abilities may move from a melee despite not having declared it.
The exact text for turning undead is "Flee from the cleric’s presence if possible, and will not harm the cleric.", emphasis mine. If the cleric's side won initiative and the undead did not declare a fighting retreat, they do not have to flee, but cannot harm specifically the cleric (but may attack the cleric's allies). At the top of the next round, during the declare movement and spells step, they must declare a retreat (which means they cannot attack).
It is unclear if clerics who turn undead can move on their turn. Can they?
The rules are ambiguous on this. As far as I can tell, clerics do not have to declare that they're turning beforehand, and it's up to the GM whether or not they can move.
Are there similar conditions to be met by clerics to turn undead as to cast spells (freedom, sole action, line of sight, no movement) ?
The rules do not specify any conditions on turn undead. OSE just says "Clerics can invoke the power of their deity to repel undead monsters encountered", which doesn't explicitly say that the cleric needs to be able to move or speak. No range or line of sight requirements are given either. It's been a problem at my table before.
IF I declared a fighting withdrawal or a retreat, but for some reason (friendly fire, more than 2 sides, etc.) my opponent is dead or no longer in melee by the time it's my turn to move, am I held to that declaration? Can I sit still? Am I now free to move forward, no longer being in melee?
Fighting withdraw says "The combatant moves backwards at up to half their encounter movement rate. There must be clear space behind the combatant." Note the lack of optionality. Not "the combatant may move backwards" but "moves backwards". Thus, I read that once a fighting withdraw is declared, they may move backwards up to half their move. 1 inch of backwards movement counts (effectively not moving), but forward movement does not.
If my opponent declared a retreat and won initiative, and fled, but I have enough movement rate to catch up to them, and am no longer in melee now that they fled, can I pursue? If I do, will I get the bonus to attack them?
you may pursue and you get a bonus to attack them if you do
Does not say the caster cannot be in melee, so I guess if they were already in melee they can still cast a spell. Is that so? What about magic items - are their special powers used during the spellcasting phase?
Yes, they may cast a spell in melee. The text is unclear on when magical item use takes place.
If I started my round with a bow in hand, and an enemy moved towards me into melee range and attacked me, can I now draw another weapon and make a melee attack?
The text is unclear on how dropping or drawing weapons works.
If an opponent declared they were going to flee (retreat), but for some reason, they cannot (we won initiative and a spellcaster just cast hold person in the previous phase for instance), do I still get the +2 to hit them and the negation of their shield bonus if I attack them?
Yes, declaring a withdraw gives your opponents +2 to hit you for the round.
There's a lot of ambiguity in BX combat, and I'm sorry that you didn't get a high-effort reply in your original thread. Hopefully this makes up for it.
Since there will be ambiguities and you have to make rulings, the best advice I can give is to use the framework we're given. Make rulings about whether or not turn undead needs to be declared, whether it can be disrupted, and whether there's a range/line of sight limit, and then be consistent.
When the players want to try something wonky (like swinging from chandeliers or jumping down from buildings), make a judgement call about how it fits into the structure (does it take their move action, their attack action, or move+attack like a spell? Is it slow like using a 2h weapon?), let your players know, and then ask them if they want to do it.
Finally, some other areas of ambiguity:
You must declare melee movement if you're already in melee, but does anything stop you from running right past someone if you start your turn just out of their range?
Does anything stop you from running through someone?
Do other combatants (like allies) count as cover for attacking a foe?
Hey, I was about to ask you what happened to it! What a fuckup on Reddit's part; they dug a huge trench through several big, important threads on r/osr.
Best of luck getting it back!