What are your thoughts on roll-under/roll-high mechanics?

[dragoner]

I'm interested in your opinion of roll-under/roll-high mechanics. The way they work is like this: you roll your die or dice (in the White Hack's case, for example, a d20). You want to roll equal to or under your Attribute or skill. But, within that metric, you want to roll as high as possible. So, if your Dexterity is 13, you want to roll equal to or under 13, with a 13 being the best-case scenario. It has been referred to as a Blackjack-adjacent mechanic. What I like about it is that roll-under is very easy to understand, and the roll-high aspect of it gives characters of higher skill the potential to reach levels of success that other characters can't. And there is no pesky math, for those who don't like adding modifiers. I've been playing around with it for my own game, using 2d10 instead of 1d20. One of the dice is the Alpha Die. That die determines your base damage if your attack is successful in combat...meaning that a character with higher skill can do more damage than someone with lower skill. What are your thoughts on the mechanic in general?

I haven't used it, sounds interesting. In my game I use 2d6 vs 8 plus mods so that PC's are usually +1 or more as I want them rolling with a 68% or 2/3rds chance or more of success. I have read that casinos like doing these odds as well as someone feels it is fair, or happier rolling. Someone on reddit became upset with me when I pointed out to them that their balanced 10 50/50 chance rolls by the player is a loss for the player, likely more than once. I don't know the odds on the Blackjack adjacent mechanic off hand so I would want to research it more.

 

[Autumnal]

I’m warming to it gradually. It’s not my favorite, but it works, and when it’s part of an overall good system, I don’t really have significant complaints.

 

[Haiku Elvis]

I like it I don't have an issue whether a game is roll over or under as long as there is a consistency through the game.

As has already been said roll under but high can help adjudicate degrees of success and opposed rolls without the maths of roll under half for extra success etc.

It can also lend itself to pushing mechanics as in blackjack itself. Twist and risk going bust for a better success or stick.

Finally it naturally works with ability scores as target numbers so reduces the pressure to come up with arbitrary but also accurate TNs all the time

 

[TheAlkaizer]

The 'Speedster' superhero should always get the drop on everyone else. The 'legendary gunslinger' PC should always outdraw his opponent. I'm not knocking dice rolls, but they should never negate a character's special abilities.

That's a bit of a drastic example. If the gap between your fastest character and the second fastest is the difference between the Flash and the Joker, then yeah, I don't see why we're rolling. However, in most games (including D&D) the gap is much more narrower. The odds are absolutely in the favor of whoever has the best bonus, but the dice represents the alea of life (context, luck, surprise, or anything) else that can happen. The "best" team sometimes loses in sports.

This is why trad d20 is so popular: it helps characters shine, especially when they gain bigger bonuses from advancement and equipment.

I always had the opposite perspective and saw this exact critic of D20. It's very swingy and the majority of the odds lie in the dice and not in your bonus. Where systems like 2D6 for example, have a different curve that makes the dice much more predictable, and so the bonus much more reliable too.

 

[Staffan]

I haven't used it, sounds interesting. In my game I use 2d6 vs 8 plus mods so that PC's are usually +1 or more as I want them rolling with a 68% or 2/3rds chance or more of success. I have read that casinos like doing these odds as well as someone feels it is fair, or happier rolling. Someone on reddit became upset with me when I pointed out to them that their balanced 10 50/50 chance rolls by the player is a loss for the player, likely more than once. I don't know the odds on the Blackjack adjacent mechanic off hand so I would want to research it more.

Blackjack mechanics, when using a single die (d100 counts as single in this case) to create a uniform distribution, is basically just a simplification of using margins of success where you don't need to do any subtraction. For example, let's say you're using a d20-based roll-under system where one character with relevant skill 15 is pitted against another with skill 12. Using margins of success, we get the following matrix based on rolls:

If we instead use Blackjack rolls, we get the following:

So you get the same number of wins and losses for either part, it's just a matter of which combinations of rolls go where. And if you instead use d20+X, higher wins, you get:

This is not the same because there aren't any failed rolls, and in the other two a failure vs failure is a tie. But otherwise it's fairly similar. The main difference is that Blackjack rolls get wonky if/when the potential skills get bigger than the die, whereas the arithmetic ones (die plus bonus high wins, or margin of success) can handle arbitrarily large numbers. Blackjack's main advantage is that you don't need to do any arithmetic: just compare numbers, first to your skill value to see if it's a success in the first place, and then to your opponent's number to see which is bigger.

Of course, if you're not using a single die but instead multiples creating a triangular or bell curve distribution, these comparisons go totally out the window. In those cases you probably should not use Blackjack rolls.

 

[Umbran]

What are your thoughts on the mechanic in general?

It sounds like a solution in search of a problem. It is more complicated than many other mechanics, but I don't see what payoff we are supposed to get for that added complication.

You don't say what genre you think will be well-served by such a mechanic, of what issue(s) you perceive with other mechanics that this is supposed to fix.

Specifically - the interesting point of blackjack is that it is a "push your luck" game. The player has decision points of whether to take another card or not. What decisions do players have in this rolling mechanism?

 

[schneeland]

I don't think it works well for games that veer into the mythic/superheroic, e.g. a game where a "high-level" warrior could be expected to fight off an entire platoon of normal soldiers by themselves, or to suplex a bear. For games like that, I think open-ended die+modifiers above TN work better.

I generally like the mechanic, but I tend to agree - while you can do additional adjustments, I feel roll-under generally works better if stats for PCs and NPCs/monsters are in the same range.

@vivsavage : If you are not already familiar with them, it might be worth checking out Pendragon and maybe also Whitehack as examples of systems with d20 roll-under/black jack mechanics - maybe there's some things you find useful for your system in there.

 

[Thomas Shey]

I’m warming to it gradually. It’s not my favorite, but it works, and when it’s part of an overall good system, I don’t really have significant complaints.

It takes some getting used to if you're used to conventional roll-low, as a group I ran Mythras for who were long-time RQ/BRP players showed, but I think it serves some good purpose. It also allows some useful application of the range space that might otherwise feel odd, as Eclipse Phase shows.

 

[Bill Zebub]

I understand the elegance/simplicity of roll-under or roll-low...but I just don't like it.

Thought experiment: imagine a roll-above game where lower ability scores are better. Same math as some roll-under games, but....low ability scores? It's just wrong.

 

[Thomas Shey]

I always had the opposite perspective and saw this exact critic of D20. It's very swingy and the majority of the odds lie in the dice and not in your bonus. Where systems like 2D6 for example, have a different curve that makes the dice much more predictable, and so the bonus much more reliable too.

Honestly, its two different issues being conflated anyway; wide die range systems (D20 or D100 both land in this) tend to produce very wide variance barring other design elements, but at the other end whether having big gaps between potentially competitive (in the sense they'll be doing opposed rolls or equivalent) opponents is desirable is a separate question; its to be noted that most superhero games do not do that (either via non-representative numbers based on story function, or by one form of compression or another (often powers-of-two statistics)) so its not considered as a particularly big virtue for such gaps to be there in the first place in many deisgns.