In my previous entry, I put forward an idea for an Attribute/Skill/Talent dice pool system. Some friendly folks on Reddit pointed out some flaws with the idea (big thanks in particular to Cryptwood and Dimirag who gave some very helpful, very constructive criticism), including:
- Generating Attributes by rolling a d20 is very swingy--you might get a 1, you might get a 20, and so it might be better to let players decide their Attributes themselves, possibly with a point-buy system. Which would be very Sims-like. I like the idea of homaging my influences like that!
- If the goal is to get as many dice over the target number as possible, using your fifth-highest Attribute (with a d4) might be seen as more preferable, as it'll allow you to get more d8s over the line than you would with just a d20. Use more Talents, get more Applause.
- Using fewer Talents with a low d4 will ensure that you'll have a harder time succeeding, as your Attribute roll will be lower.
- With a d20, you only need one or two Talents realistically to get over the line, and you'll succeed more often than not, but you won't get as much applause unless you use a LOT of Talents.
- In theory, you'd be using the same number of Talents in both cases--why wouldn't you want to maximise your dice?
- Attribute names were hard to understand--I don't disagree--but again, they're not final! I can't stress that enough, the Attributes I gave were only examples, and subject to change in the final version.
Using a d20 (for your highest Attribute), plus 1d8 (for your Skill) and 5d8 (for your Talents) nets you a bell curve like this:
If we assume average rolls across the board, that's 10 (1d20), 4 (1d8), 4 (1d8), 4 (1d8), 4 (1d8), 4 (1d8), 4 (1d8), which equals 34. The percentages when rolled together tend to skew a bit higher; 37 and 38 are equally likely (at 5% apiece), while a 34 is only 4% likely. So, let's assume 4 of those d8s rolled fives instead: 10 (1d20), 5 (1d8), 5 (1d8), 5 (1d8), 5 (1d8), 4 (1d8), 4 (1d8), for a total of 38.
If the target number is, let's say 17, the third die (1d8) gets you to 18, and in excess of the TN. So, you'd get 5 Applause.
Meanwhile, 1d4 + 6d8 looks like this:
Assuming the same TN of 17, the first 4 dice get you to exactly 17; so you'd only get 3 Applause. That doesn't seem so bad! A lower Attribute gets you less Applause. I'm not so concerned about that.
Now, it's possible that some player psychology will come into play: players will see that d4, and instinctively want to use MORE dice to improve their chances of success. Is that a bad thing? I'm not entirely convinced. Not every Talent will be useful in EVERY situation (remember, Talents are quite specific, it's Skills that are broad--"parallel parking" versus "driving"), so it's possible that this problem will take care of itself.
I think a major point of contention is that I will have to ensure that advice is given to GMs in the rulebook that Talents should be extremely specific, so as to avoid use cases where a player crams every Talent they have into a dice pool. Not every Talent should be useful in every situation!
However, there's one aspect to all of this that I hadn't considered, and I think it's arguably the most important bit of critique:
Counting huge dice pools takes a long time. It'll slow down play dramatically, and that's why many dice pool-based games instead count successes. This, however, makes target numbers based on Attributes completely superfluous, so if I intend to stick with dice pools, I need to come up with a new resolution mechanic if I don't want play time getting bogged down by slow players.
Comments
Post a Comment