Anti-Patterns in TTRPGs: Glacially Slow Games, Bloated Mechanics, and Poor Abstraction Scaling.
Or "Hey! Put your phone away!"
This time I want to cover the anti-pattern “Glacially Slow Games, Bloated Mechanics, and Poor Abstraction Scaling” This topic was motivated by an exchange I had on X/Twitter in June with
and about combat grids, types of combat turns, freeing player imaginations from thinking in videogame-like ways, and eventually abstraction levels. I have been mulling over that discussion a lot since then, and I don’t think I expressed myself very well at that time, nor were my thoughts properly well formed at that time. I think can better explain it this way by talking about what makes TTRPGs slow, and why the ability scale levels of abstraction as needed or wanted is important to all the above.As a refresher allow me to give a definition of an Anti-Pattern:
“a common, well-intentioned response to a recurring problem that is at best ineffective and usually highly counterproductive.”
Overlong Actions and Bored Players
We’ve all been there with some TTRPG, its been 20+ minutes since you last had anything to do. Maybe the Wizard/Magic-User is trying to pick a spell, maybe combat is just slow, or maybe some player has been negotiating with or spewing a monologue at an NPC by himself since he has by far the best social skill ranks.
Either way by now everyone except the GM and one player is on their phones doomscrolling. That is not fun for most people at the table, and absolutely no one thinks this is fun when they are one of the “waiting ones.”
This is not a even a game at this point. If it was a sporting even this would be called “halftime” and most people would have gotten up to grill some burgers, or hotdogs, or drink a couple beers, or do literally anything but sit there with their thumbs up their butts for 20+ minutes. Even chess has time limits. Baseball now has pitch clocks to keep the game moving.
Yet, modern TTRPGs are painfully slow and do nothing about it.
How did this happen? Why does this happen?
Well like most issues in this hobby, its wasn’t intentional. In this case it is actually the accretion of a number of small decisions that ended up causing big problems by the end. Much like how someone gets fat, this outcome is not the result of a single bad decision.
So what is the problem? How is this an anti-pattern. What issues are game designers and GMs trying to solve? Well… its a long list, but every item on it from long combat turns to monolguing players basically boils down to game designers, GMs and whole game systems trying to make the game better by making some internal system more fleshed out, but doing so in a way that does slows the game down, and doing so without providing a way to increase the level of abstraction if needed/wanted to do the same task quickly. This eventually drains the fun out of the game from pure slowness. That highly counter-productive, hence anti-pattern.
The Road to Hell is Paved With Good Intentions
Allow me to expound via an example:
Having lots of combat options and unique mechanics can be fun for small fight. Its nice to be able to trip, grapple, target specific bodyparts of enemies, etc. But that all takes time to resolve. What happens in medium and large fights?
As an example, rolling for individual initiative works fine when there are merely 6 orcs attacking the 4 PCs. But what about 16? Or 36? (or in AD&D 200?) Now what? Are you really going to roll 36 d20s, write them all down on a list along with which orc it belonged to, then add the requisite modifiers, and then sort them all? Just to run a single combat turn?
Oh wait, that didn’t even account for the PCs.
What about needing to have those 36 (or 200) orcs make a saving throw to not slip and fall on a newly created oil slick the magic user just made while they were charging? Are you really going to roll 36 saving throws?
On a different note, do I really need to run EVERY SEGMENT of EVERY ROUND in AD&D? Really? Are you sure that is what was intended?
How about a combat grid? Do I ALWAYS need a combat grid to track the movements and relative positions of PC and NPCs/enemies? Or only at certain times? Does the grid need to be “On” or “Off” for a whole battle?
And yet you say, having massive combats that alter the fate of cities and even whole civilizations is fun! Having lots of character options is fun! Having a lot of useful and unique spells is fun!
But what if those fun options slow the game down? I have mentioned
’s concept of T.U.N.I.C. (Time Until Next Impactful Choice) before, but it bears re-mentioning here, as it is a useful metric. Things that increase the T.U.N.I.C. past a couple of minutes can really suck the fun out of the game.(EDIT: RuleOfThule made an excellent point in the comments that T.U.N.I.C. simply does not apply in a number of situations, and I found it convincing. Give it a read there or I have the relevant part reprinted here1)
Ok so, can we have our cake and eat it too? Can we have these things and not have a game the flows like tar in an Alaskan winter?
Yes you mostly can. But the game is going to look quite different.
That’s because the real issue here is:
Missing Mechanics and Lack of Abstraction Scaling
You must understand, every single mechanic in a TTRPG is an abstraction of some event or action in the game world. Travel, swinging a sword, buying an item, literally everything anything is an abstraction instantiated into the game with rules or rulings (hopefully precedent generating rulings) to simulate SOMETHING.
Abstraction Scaling is when you alter the level of abstraction being used to simulate the game world to better fit the situation. Even something as simple using different mechanics to move in different situations is abstraction scaling. Consider how movement is resolved differently when:
Traversing a dungeon;
Repositioning in combat.
Travel overland between cities.
Travel inside a peaceful town.
No one is going to play out traveling 200 miles to the nearest big city on a combat grid with 1inx1in squares representing 5ft, and having everyone in the party to move tokens/minis across the grid in initiative order in 6 second turns (at least I hope not). Instead they increase the abstraction level and use hex crawling rules with a scale of X miles per hex and multiple hours per unit of time (or abstract further and use point crawling rules, though I don’t recommend this).
That is an example of abstraction scaling. The level of abstraction of the simulation was changed. In this case you can also think of it as “changing game modes” in the TTRPG (combat to overland travel); either way the nature of the simulation was changed.
Similarly if there is an encounter while traveling the abstraction level is dropped back down so the players and PCs can interact with whatever it is on a more real-time level (talking to merchants, fighting a band of Ogres, etc).
Again absolutely everything is a TTRPG is an abstraction to simulate something.
Do we NEED to have the PCs roleplaying out their haggling with a shopkeeper to buy rope? No you don’t have to. You don’t have to let the game turn into a simulated shopping mall run. That can be abstracted to just the PC saying “I go buy some rope before we set out in the morning.” Maybe add an opposed charisma or intelligence check to simulate haggling. You also can run the whole interaction. Each method is a different level of abstraction.
If a PC is arrested in a corrupt society where they have a 1/100,000 chance of acquittal at trial despite being innocent, do you HAVE TO to run the trial? No you don’t HAVE TO; you can if you want, but you don’t absolutely need to. You COULD just raise the abstraction level to resolve the trial as a single roll, and let the players get on with planning a jailbreak or trial by combat or whatever. Its your choice as GM as to what level of abstraction you scale up or down to, almost whenever you want.
These are reduction to absurdity examples. There are many many more. This type of scaling can be applied to literally anything in a TTRPG.
My point is that one of the main ways the game becomes glacially slow is game systems not helping the GM to scale the level of abstraction as needed to keep the game moving. Most of time with d20 type games this happens in combat.
Going back to my earlier example of fighting 36 orcs; what that game needs is a way to abstract and simplify imitative, or combat as a whole for that matter, so that fighting 36 orcs could be treated as fighting 6 orcs or even 1 orc.
Those mechanics for scaling the game as needed are simply not there in most systems.
Sure, some problems are that cause slowness are not related to abstraction scaling. The problem with players taking forever to look up spells is due to a lack of a game mechanic that states something like: “At the start of a combat round you must IMMEDIATELY declare any spells any PCs of yours are going to cast or they can’t cast any spells that round.” Harsh, but useful.
But that second one is a small issue, see here2 for slightly longer commentary on it. A lack of proper abstraction scaling is the main problem most of the time. The ability make the game’s simulation “zoom-in” or “zoom-out” to resolve things with more or less detail and time spent is crucial to a game running well. The best examples I can think of for Abstraction Scaling working well are AD&D’s assassination mechanic, and the “Chainmail” inspired combat scaling
’s “Hellmarch.”My AD&D assassination example comes from
’s post here:
He explains it in more depth than I will.
Basically Assassins in AD&D have an ability to use a very high level of abstraction when they try to assassinate someone. If they can surprise someone, or even show they have a relatively foolproof plan to surprise someone, they can just roll on a table to see if an entire assassination works or not. In many cases this can literally be a a couple roles of 1d100. Yet at the same time you can also plan the whole hit and make the extra roles for smaller things like disguise checks, hide checks, etc.
The whole hit was abstracted away into a couple d100 rolls and say 2 minutes of game time.
THAT is abstraction scaling at work. You CAN run the whole assassination attempt in gory, gritty detail of how the assassin comes upon his prey unawares, deals with bodyguards, the fight and blow by blow attacks and damage of said fight… OR you CAN just roll 1d100 1-3 times (disguise check, assassinate check, and some other check for the escape) and know the answer in 30 seconds. The second way is far more abstract, and thus much quicker than the first.
You can get even more abstract. In AD&D you can just hire an NPC assassin and roll 1d100, once from the assassination table. Reasoning from R.A.W. you take an unspecified SMALL penalty for not helping plan the heist (due to not planning mitigation for any precautions the target may be taking), but regardless of that, it’s literally Pass/Fail.
This is abstraction scaling at work. We can go from using many small abstractions to simulate small actions that are part of a assassination “hit,” such as attempts to hide, attack, damage, disguise, etc all the way to using a single large abstraction that simulates the entire “hit.”
What of my other example Hellmarch/Chainmail?
In those systems there are simple rules to combine multiple instances of roughly the same creature (an orc, or a human guard) into a single squad type unit, that can interact with other squad type units and also interact with and contain hero type (PC) units.
It’s magical. In one session 4 hour session I’ve run combats with 160 giant skeletons, a evil Lich, and hundreds of human troops complete with very high level Magic Users and Clerics in Hellmarch.
Even with a bunch of player skill related slow downs, IT DIDN’T EVEN TAKE A THIRD THE SESSION. We ran a dungeon crawl after that in the same 4-hour session. That was 300+ unit combat with fancy magic in ~1 hour.
How you ask? Hellmarch’s mass combat rules allowed me to used 20-1 scaling for the skeletons, and 40-1 for the humans. It also allowed me to simplify how attacks were run in the combat. As a result it was more like running 8 skeletons and 8 humans, (plus the high level PCs and NPCs). Boom manageable fight.
That is abstraction scaling at work. I didn’t need to simulate/run every single NPC, I could combine them into higher level abstractions and run the game much faster.
By now you should be able to see how allowing for proper abstraction scaling can solve most game slowdown issues and solve almost all of “Glacially Slow Gameplay” issues.
How to run AD&D’s segments demonstrates another form of abstraction scaling. You see, most of the segments having nothing of note happening in the unless someone is moving around or trying to take an unusual action like retrieve and drink a potion or cast a spell. Thus unless people are moving around in the combat or taking unusual actions, there is no need to run all the segments. The Blue Bard has a good write-up on this. If the battle formations have engaged each other and are just trading blows, that is all you need to simulate. That is literally one segment of action per turn per side. You can abstract all the other segments away until one side breaks and runs or cavalry come rushing over the hill to break it all up. So scale up the level of abstraction to account for it, and just don’t bother simulating 9/10 segments per round.
Similarly when something interesting does happen, scale the abstraction level back down to running most or all the segments as needed to simulate that interesting bit. This is abstraction scaling working in the other direction.
For example, one time my players were caught in a relatively wide hallway (~40’) in an Appendix A procedurally generated dungeon with a large amount of something nasty fighting them (I don’t recall what exactly). They decided to use the ability to move every segment to make a fighting retreat into a defensible room with a 10’ door and fight there. So I ran every segment of their retreat in gory detail because EXACT positioning mattered. Who was in range of the enemy? Who go through the door when? Was anyone flanked? All this mattered and was based on EXACTLY HOW the players went about this tactical retreat.
BUT, once they were in the room it was just trading blows between 2 PCs and 4 whatevers, so I went back to abstracting away 9/10 segments. This is also abstraction scaling.
Sticking with combat for a minute, consider the fact that unless you need to track EXACT positioning, you don’t actually need to use a combat grid. By this I mean flanking, back attacks, line of sight for archers/spells, distance for charges, etc. For an example of when you do not need a grid, consider the above situation where the PCs are using a 2 square wide hallway in a dungeon to fight off a pack of whatevers. At any given time only 2 PCs can melee against 2-4 whatevers. There is no chance for anything but a small melee and maybe a spells like magic missile or sleep that again do not need exact positioning. So in that case don’t bother with the grid, just scale up the abstraction level and use “Theater of the Mind” since the detailed simulation gains the game nothing here. As noted earlier you can and should also abstract away the segments in that situation.
Now lets consider initiative. It would be nice to be able to use individual initiative for VERY small combats (2v2 or less) and side based initiative for everything else. I know of no system that allows this, but there is no reason it can’t be done.
I much prefer side based initiative for my game because if you account for mercenaries and personal armies, I have not had a combat with the PC side being less than 25 units in many, many months (usually more like 60-80). Accomplishing initiative in 2 rolls ( 1 roll per side ) saves a lot of time.
An example brought up in my by
twitter exchange that motivated this post involves the higher level abstraction of that results from side based initiative combined with AD&D’s 1-minute rounds. The idea that with 5 PCs vs 25 orcs the GM can literally just throw 5 orcs at each PC for a grapple check, and just throw 5d20 at each PC. No need to resolve each orc’s grapple attempt sequentially, they all get resolved simultaneously. With individual initiative you would be forced to resolve these attacks one at a time, and maybe even shift who grapples who based on how quickly a given PC fails the grapple check. To me this does not require a 1-minute round, so long as the orcs could close to melee range during the round. Either way this shows the fact that some things are better simulated at higher abstraction levels. (In theory with individual initiative you could delay some of the orcs movements so they all moved in packs of 5, but you would be fighting the game-system and taking penalties in the form of extra PC actions to do so. )For, another example what if the PCs are high level? Shouldn’t the game has a way to abstract encounters where the outcome is literally guaranteed to be a “flawless victory” for the PCs? What if some party of roughly 10th level PCs AND THEIR ARMY run into a low level encounter, say 10 wolves? Are you really going to run that? Of course not. You are going to say the wolves run away, or the archers make pin cushions out of them or some similar thing. This is an ad-hoc form of abstraction scaling. You abstracted the combat into a rough chance of anything meaningful happening and decided whether to simulate it or not based on reasoning at the highly abstract level.
Recall this doesn’t only apply to combat. Social interactions with NPCs that you the GM choose to resolve without role-playing out the whole thing is also abstraction scaling. Something as simple as you allowing player to ask “Hey GM, what rumors are floating around town today?” Is abstraction scaling as you are not making the go roleplay to some NPC for that information. So is not roleplaying the guards of a castle, but roleplaying the conversation with the King who lives there. Once you understand that you the GM control the granularity of the simulation, you can find uses for this concept of abstraction scaling everywhere.
In conclusion literally everything in a TTRPG is an abstraction instantiated into the game with rules, rulings, or more abstract things like roleplaying. Abstraction Scaling is when you alter the level of abstraction being used to simulate the game world to better fit the situation. Systems that do not help the GM scale the abstraction level of the simulation are dooming their game to be slow and unsatisfying.
Like what I have to say? Hate it? Either way:
You can also follow me on X/Twitter @thalassiosx2
“In a spotlight-sharing mono-party adventure where we imagine players are rifling “live” through the loot in dungeon, this notion makes a lot of sense; after all, we want to spread that spotlight around evenly and judiciously.
However, in modes of play outside this narrow scope, the landscape of decisions & consequences is messy, highly non-linear, and fraught with more peril and meaning. For example, consider the process of planning for a defensive skirmish. We must assemble units, take into account placement & positioning, provide for safe lanes for supply and/or VIPs, cover/control key terrain, and so on. All of these decisions are made in an abstract “planning” time, unconcerned with spreading spotlight to any players.
And when the time comes where we are running the battle “live”, there is a question of what orders to submit. The referee sees them, rolls initiative, and we “merely” find out what happens, at length! But this can be 100x as engaging, nerve-wracking, and campaign-defining than “deciding” to backstab a guy within 15 seconds or less. And the temporal distance between decision points can be very long! Borrowing Beast’s heuristic does not account for the joy of finding out or the satisfaction of a well-laid or -executed plan. If we were to follow his design guidance, we would ruthlessly eliminate these things from our game’s structure.“
—Excerpted from RuleOfThule in the comments
This is an issue I call “All options, no stick,” which is giving the players lots of options and bonuses with no MECHANICAL incentive to make quick decisions and stay on top of the math. It is a great way of making the game take multiple sessions to get through a single combat turn.
Consider time players spend adding combat modifiers from their abilities or spells cast on them, every damn turn? Why won’t the players just pre-calculate this stuff? So make a rule, pre calculate your bonuses, or they don’t apply.
What about players getting stuck trying to pick a spell? How many HOURS of your life did you pour down the drain waiting for somebody else to look up or pick out a spell?
This is a “missing rules needed to keep the game moving” problem. So make some such as the aforementioned “declare any spell-casting when the round starts, even before initiative or lose the ability to cast that round.”
In short players. will take forever to do things unless there is a mechanical reason not to take forever. So give them a mechanical reason.
Great piece; well-rounded and directly attacks all the important angles of the concepts presented.
A (very minor) critique:
"I have mentioned Questing Beast’s concept of T.U.N.I.C. (Time Until Next Impactful Choice) before, but it bears re-mentioning here, as it is a useful metric."
It is with great pleasure that I countersignal the Borrowing Beast here on this point. In a spotlight-sharing mono-party adventure where we imagine players are rifling "live" through the loot in dungeon, this notion makes a lot of sense; after all, we want to spread that spotlight around evenly and judiciously.
However, in modes of play outside this narrow scope, the landscape of decisions & consequences is messy, highly non-linear, and fraught with more peril and meaning. For example, consider the process of planning for a defensive skirmish. We must assemble units, take into account placement & positioning, provide for safe lanes for supply and/or VIPs, cover/control key terrain, and so on. All of these decisions are made in an abstract "planning" time, unconcerned with spreading spotlight to any players.
And when the time comes where we are running the battle "live", there is a question of what orders to submit. The referee sees them, rolls initiative, and we "merely" find out what happens, at length! But this can be 100x as engaging, nerve-wracking, and campaign-defining than "deciding" to backstab a guy within 15 seconds or less. And the temporal distance between decision points can be very long! Borrowing Beast's heuristic does not account for the joy of finding out or the satisfaction of a well-laid or -executed plan. If we were to follow his design guidance, we would ruthlessly eliminate these things from our game's structure.
While I tend to agree with the general gist of your argument, I also have to ask, Why is the goal to make everything happen fast. If players get bored waiting around, that’s a player issue, not necessarily a mechanics issue.
If I had spent years of game time amassing an army to fight the evil orc horde, it would be pretty anti-climactic for the battle to be over on 10 rolls. I want that to be an epic battle that maybe last two sessions.