Tabletop Arcade Dev Diaries #15: Advantage and Disadvantage
Hey everyone! In today's dev diary I'm gonna take a quick look at different advantage systems in tabletop games and settle on an advantage system for Tabletop Arcade.
When I talk about an advantage system here, I mean the system that translates circumstantial advantages to solid mechanical bonuses a character gets to a particular roll. The keyword here is circumstantial, characters only gain these bonuses temporarily, to particular rolls, or under particular conditions.
There is clearly a lot of variety in these systems, so I want to lay out the goals of this system in Tabletop Arcade. For one, it should be simple to use. All else being equal, cutting down on the time it takes to do admin at the table is definitely a plus. I also want to preserve bounded accuracy - in summary, you shouldn't be able to stack advantages over and over again until the results of your roll are as high as you want.
It may sound strange that I want to promote ease of use in TA after the ability construction draft chapter I released last week. How can someone that wants you to file a tax return to make an ability say they want ease of use? Well I really do want the game to be easy to play, and if you'll forgive the detour I'd like to explain myself just a little. Personally, I think when it comes to tabletop games, you can divide complexity into two classes. The first I'm calling "backstage complexity", and this is anything that you do in preparation of actual play. On the player's side, this is mostly creating your character, choosing equipment, and in this case creating or choosing your abilities. The second is "performance complexity" - in a nutshell, how difficult is it to translate your intended action to a die roll when playing?
When it comes to TA, I'm doing my best to avoid this latter form of complexity. I'm significantly less worried about backstage complexity, for a couple reasons. For one, players can get help from each other and the GM without much issue during preparation. Also, you can avoid it by simply providing pre-made resources, which I certainly intend to do when it comes to the customisable abilities. All in all, I tend to think performance complexity is usually harder to avoid, and avoiding it usually means missing out on something, so I want to focus my attention on reducing it far more than backstage complexity.
Wow that sure was a bit of a ramble. Let's get back on topic, shall we? I'm going to go through a couple of the advantage systems I've seen, and try weighing up their pros and cons when it comes to how well they fit in TA.
Now that's not to say this type of system is always doomed for failure. It can work well in games if your players really enjoy more "tactical" combat and optimising strategies. I've personally never played Pathfinder, but that certainly seems to be at least part of the appeal. This type of system has also been used to good effect in Monster of the Week, and despite being essentially the same system it functions pretty differently to in Pathfinder. Monster of the Week uses a 3d6 roll to determine action outcomes, so results are naturally clustered near 10 rather than near extremes. As such, there is a pretty narrow numerical window between success and failure, so a single +1 can increase your chances of full success by about 12%. In short, small bonuses are more significant because of the core dice roll being used.
Now seems like a good time to once again reference The Angry GM, this time their article "Ask Angry: Advantage Stacking and First-Time Campaign Building". The idea I want to focus on is essentially this: if your advantages stack, players will often chase after the biggest bonus they can get, even if it doesn't necessarily confer that much benefit when it comes to the actual probability of success. This is something I'll come back to later, but it is particularly important for looking at this "bonus die" advantage system. The situation I mentioned previously - a character stacking a bunch of advantage and only ending up with a small bonus - won't be particularly rare. And on the flipside, the Gm will regularly have to deal with players getting pretty high bonuses at random.
Open Legend has a very similar advantage system, but with a few tweaks. Characters in Open Legend don't receive static bonuses from their attributes, instead they roll extra dice. Advantage in Open Legend normally applies to these extra dice, and characters can stack multiple levels. Having multiple levels of advantage means you roll more and more dice, and only use the highest values. In the case that you don't have an attribute bonus and you're rolling only a single d20 the advantage system is actually identical to 5th edition. No advantage stacking, just one extra die and pick the highest.
This sort of advantage system is the main contender for Tabletop Arcade. I was curious about tweaking it so advantage would stack, and was pretty curious about why stacking was disallowed on unmodified d20 rolls in Open Legend despite being permitted on attribute dice. This is what lead me to the Angry GM article I mentioned earlier. It seems like the main opposition to this is psychological rather than mathematical. Stacking advantage would certainly exhibit diminishing marginal utility, but most people seem to think that players would still chase after the highest advantage levels possible. It also runs into issues with a niche build using critical hits. The 5e "Champion" fighter subclass expands the range of numbers that can score critical hits from 20 to 18-20. If you stack, say, 3 levels of advantage with this build, your chance of scoring a critical hit is almost 50% (versus a chance of 27.75% under normal advantage). This is pretty clearly more than a little broken. A pretty niche issue to be sure, but avoiding it here is probably for the best. So to summarise, stacking advantage is a bit of a no-go.
When I talk about an advantage system here, I mean the system that translates circumstantial advantages to solid mechanical bonuses a character gets to a particular roll. The keyword here is circumstantial, characters only gain these bonuses temporarily, to particular rolls, or under particular conditions.
There is clearly a lot of variety in these systems, so I want to lay out the goals of this system in Tabletop Arcade. For one, it should be simple to use. All else being equal, cutting down on the time it takes to do admin at the table is definitely a plus. I also want to preserve bounded accuracy - in summary, you shouldn't be able to stack advantages over and over again until the results of your roll are as high as you want.
It may sound strange that I want to promote ease of use in TA after the ability construction draft chapter I released last week. How can someone that wants you to file a tax return to make an ability say they want ease of use? Well I really do want the game to be easy to play, and if you'll forgive the detour I'd like to explain myself just a little. Personally, I think when it comes to tabletop games, you can divide complexity into two classes. The first I'm calling "backstage complexity", and this is anything that you do in preparation of actual play. On the player's side, this is mostly creating your character, choosing equipment, and in this case creating or choosing your abilities. The second is "performance complexity" - in a nutshell, how difficult is it to translate your intended action to a die roll when playing?
When it comes to TA, I'm doing my best to avoid this latter form of complexity. I'm significantly less worried about backstage complexity, for a couple reasons. For one, players can get help from each other and the GM without much issue during preparation. Also, you can avoid it by simply providing pre-made resources, which I certainly intend to do when it comes to the customisable abilities. All in all, I tend to think performance complexity is usually harder to avoid, and avoiding it usually means missing out on something, so I want to focus my attention on reducing it far more than backstage complexity.
Wow that sure was a bit of a ramble. Let's get back on topic, shall we? I'm going to go through a couple of the advantage systems I've seen, and try weighing up their pros and cons when it comes to how well they fit in TA.
Floating Modifiers (D&D 4e/Pathfinder)
Personally, I think that this is the most obvious advantage system. Get a flat modifier to the roll based on the level of advantage. It's pretty easy to understand, and you can also contextualise the bonus pretty well by relating it to your attribute bonuses. But it has some pretty clear flaws, and really isn't what I want to use in TA. For one, it generally doesn't work great unless you let the bonuses stack. A single +1 bonus to a d20 roll isn't really significant by itself. But then it can become pretty cumbersome to keep track of lots of small, separate bonuses, and succeeding that you still have to sum them all together which can contribute to the game slowing down.Now that's not to say this type of system is always doomed for failure. It can work well in games if your players really enjoy more "tactical" combat and optimising strategies. I've personally never played Pathfinder, but that certainly seems to be at least part of the appeal. This type of system has also been used to good effect in Monster of the Week, and despite being essentially the same system it functions pretty differently to in Pathfinder. Monster of the Week uses a 3d6 roll to determine action outcomes, so results are naturally clustered near 10 rather than near extremes. As such, there is a pretty narrow numerical window between success and failure, so a single +1 can increase your chances of full success by about 12%. In short, small bonuses are more significant because of the core dice roll being used.
Bonus Die
This isn't actually an advantage system I've seen used anywhere, just a small idea I considered for TA. Basically, you'd have an advantage level granted by the various in-game circumstances, and you would roll an extra die based on this level, adding the result to the roll. So with advantage 1 you'd roll a d20 and a d4, adding the results together, whereas on advantage 2 you'd roll a d20 and a d6. Or something along those lines anyway. I initially thought this seemed pretty cool. After all, rolling dice can be fun, it might add some tension when you realise the result all hinges on the advantage die, and there's a natural stopping point at d12 to avoid stacking extra advantage on top. But having taken a closer look at the maths (thanks AnyDice) it just seems way too swingy. It'd feel pretty bad if you stack as much advantage as possible and still only end up with a +1 to show for it.Now seems like a good time to once again reference The Angry GM, this time their article "Ask Angry: Advantage Stacking and First-Time Campaign Building". The idea I want to focus on is essentially this: if your advantages stack, players will often chase after the biggest bonus they can get, even if it doesn't necessarily confer that much benefit when it comes to the actual probability of success. This is something I'll come back to later, but it is particularly important for looking at this "bonus die" advantage system. The situation I mentioned previously - a character stacking a bunch of advantage and only ending up with a small bonus - won't be particularly rare. And on the flipside, the Gm will regularly have to deal with players getting pretty high bonuses at random.
Best Bonus Die (Shadow of the Demon Lord)
Shadow of the Demon Lord uses an interesting system, that seems like a refinement of the "bonus die" idea. Essentially, you roll a number of d6's equal to your advantage level, and take the highest result as the bonus. This has a pretty similar vibe to the previous system, but deftly solves a lot of the concerns I had with it. It's far less variable, and gaining higher advantage levels actually make the benefits more consistent rather than less. A nice system, definitely a contender for TA.
As a side note, it's probably worth noting that SotDL also uses floating modifiers on top of this system, though these seem to be significantly rarer. There's some merit to having multiple advantage systems in a game, and I guess technically even D&D 5e has multiple different ways to improve your rolls. The obvious example to me is Bardic Inspiration, but there's plenty of other effects like that throughout the game. I reckon the main reasons that sorta thing is included are to just adds a little bit more tactical crunch, and to avoid the constant monotony of so many abilities just granting advantage.
This advantage system also neatly illustrates another idea I'd like to touch on, that being "diminishing marginal utility". It's a term I'm borrowing from economics no wait don't leave I have a point I swear. In this context, diminishing marginal utility is the idea that the benefit of gaining one more level of advantage becomes smaller as you gain more and more advantages. Gaining the first advantage is a big boost, gaining the fifth is only a minor improvement from before. In the SotDL system, the average numerical boost of one advantage is +3.5, but the average boost of two advantages is +4.47. It should be pretty easy to see that the maximum average boost you can get here is +6, since that's the highest number on a d6. This idea of diminishing marginal utility is key to meeting the design criteria I have for TA, in particular the desire that infinitely stacking advantages to get arbitrarily high bonuses isn't possible. Also, I think it just makes the game a bit more interesting in a couple ways. If you have the opportunity to grant a level of advantage, it's better to give that to someone who doesn't have any advantage yet, so teamwork is promoted. Once everyone has just a couple advantage levels, keeping stacking more is less effective so your tactics might change. I could keep talking about this for a while but this sidebar is already getting long enough to be it's own section so I'll be moving on.
This advantage system also neatly illustrates another idea I'd like to touch on, that being "diminishing marginal utility". It's a term I'm borrowing from economics no wait don't leave I have a point I swear. In this context, diminishing marginal utility is the idea that the benefit of gaining one more level of advantage becomes smaller as you gain more and more advantages. Gaining the first advantage is a big boost, gaining the fifth is only a minor improvement from before. In the SotDL system, the average numerical boost of one advantage is +3.5, but the average boost of two advantages is +4.47. It should be pretty easy to see that the maximum average boost you can get here is +6, since that's the highest number on a d6. This idea of diminishing marginal utility is key to meeting the design criteria I have for TA, in particular the desire that infinitely stacking advantages to get arbitrarily high bonuses isn't possible. Also, I think it just makes the game a bit more interesting in a couple ways. If you have the opportunity to grant a level of advantage, it's better to give that to someone who doesn't have any advantage yet, so teamwork is promoted. Once everyone has just a couple advantage levels, keeping stacking more is less effective so your tactics might change. I could keep talking about this for a while but this sidebar is already getting long enough to be it's own section so I'll be moving on.
Rerolling Dice (D&D 5e, Open Legend)
This is how I'm describing the advantage system in D&D 5E. I'm not shy about the fact that I really, really love this advantage system It just performs so well, and can be explained in like two sentences. It interacts pretty nicely with a whole range of game mechanics, and it's a pretty significant boost without being game-breaking. It also has the feature that gaining advantage cannot allow a character to do something that would normally be impossible for them - the maximum they can roll doesn't change. This is a particular vibe that may or may not fit certain games, but I think it's pretty great here.Open Legend has a very similar advantage system, but with a few tweaks. Characters in Open Legend don't receive static bonuses from their attributes, instead they roll extra dice. Advantage in Open Legend normally applies to these extra dice, and characters can stack multiple levels. Having multiple levels of advantage means you roll more and more dice, and only use the highest values. In the case that you don't have an attribute bonus and you're rolling only a single d20 the advantage system is actually identical to 5th edition. No advantage stacking, just one extra die and pick the highest.
This sort of advantage system is the main contender for Tabletop Arcade. I was curious about tweaking it so advantage would stack, and was pretty curious about why stacking was disallowed on unmodified d20 rolls in Open Legend despite being permitted on attribute dice. This is what lead me to the Angry GM article I mentioned earlier. It seems like the main opposition to this is psychological rather than mathematical. Stacking advantage would certainly exhibit diminishing marginal utility, but most people seem to think that players would still chase after the highest advantage levels possible. It also runs into issues with a niche build using critical hits. The 5e "Champion" fighter subclass expands the range of numbers that can score critical hits from 20 to 18-20. If you stack, say, 3 levels of advantage with this build, your chance of scoring a critical hit is almost 50% (versus a chance of 27.75% under normal advantage). This is pretty clearly more than a little broken. A pretty niche issue to be sure, but avoiding it here is probably for the best. So to summarise, stacking advantage is a bit of a no-go.
In Closing
Well that sure took longer than I expected. In summary, I'm probably going to go with a simple 5e-like advantage system, though I'm considering something similar to the SotDL system as well. The 5e system has the edge for a couple reasons - ease of use; it's interactions with critical successes; SotDL disadvantage feels a little too crippling. I mostly avoided talking about disadvantage in this article to cut down on length and since most of the points are just the same, but don't believe for a second that I'm not thinking about it as well.
I'm glad that's out of the way. This article is honestly something I've been putting off a little due to the somewhat technical subject, but recently it had become a bit of a bottleneck in production so I had to get it done. I reckon next week's post is gonna be the Feats draft chapter, I hope to see you then!
Comments
Post a Comment