The core rolling mechanic in There's Glory in the Rip is the action pool. Each player has a finite set of 6-sided dice that they can use to roll for actions. Each time you roll a die, you lose it until the next refresh (effectively the next round). If you roll the max value on the die, though, you don't lose it. Players start with 3 dice in their pool, but can get up to 6 from role-playing or from trying risky things and failing.

I really like this system! On the face of it, it's kind of like a 3-action-point system like you'd find in Pathfinder 2e. That was, in fact, the original inspiration. I like that there's more of a tactile element to this one, though. You've got 3-6 physical dice sat in front of you, and as you roll each one, you set it off to the side until the next refresh. This gives you a strong visual cue for how many things you can do. I originally tried playing around with the idea that you'd roll your dice towards the GM, giving them dice to use for NPCs, but couldn't really get that to work well with more than 2 people playing.

Maybe that's an idea for another game.

The mechanic of action "points" equating to rolls, which are equated to physical dice sitting in front of you, has a lot of really interesting properties that I've tried to make use of throughout TGitR's core system:

• If dice can sit in one group in front of you, why not have more than one group? This lead to character talents that "reserve" some number of dice into a separate pool for use for specific actions, which last as long as the talent does.
• If the length of your turn is dictated by the number of rolls, is there a reason not to allow rerolling for an action? No, I decided. You can keep rolling dice (and keeping the last value) until you succeed. You just won't be able to do much else with your turn.
• Can you use up dice in your pool in a way other than rolling? That lead to abilities that activate without a roll, but require 1 or more dice to be used to pay a cost. Once you get extra dice through Glory, you can give up those extra dice to activate super-powerful abilities. More on this later.
• If a player's turn potential is related to the number of dice they have, can you balance encounters around the number of dice the players have? Turns out, yes! Seemingly! With some very simple simulations, it turns out that to get a combat with a similar overall chance of success for the players, the difficulty and number of monsters scales with the number of dice players have each turn.

One major difference between TGitR and an action-point system like Pathfinder 2e is that defending is an active roll on the player's side, done in response to NPCs telegraphing their attacks. This means you as a player have to decide how many dice you want to devote towards defending yourself vs trying to kill your target. If you can find creative actions that achieve multiple things at once, you'll be able to make your dice stretch further.

Balancing Player Talents Using Dice Efficiency

If you are using a die in your action pool to pay to active a talent, you always need to compare that ability to just using the die cost for more actions. Spending 1 die to empower your attacks and then roll 2 dice for those attacks should be as good or better than rolling 3 dice for attacks. Otherwise there's no point to using the talent! Thus, I've had to do some careful math to try to make it so that each talent is worth the dice you spend it on. This is even more important for talents with a Glory cost, which remove the extra dice you gain from Glory until you can earn more. Something that costs 1 Glory equates to losing 1 die from your pool, so it needs to be comparable to spending 1 die from your pool every turn for as long as it lasts PLUS extra to account for the lost power between the talent ending and you re-earning the Glory you spent.

Luckily there's some math you can do to figure out this balance.

By default, every action that can deal damage to something deals 1 strike. From there, you can calculate the average number of strikes dealt per die for different difficulty levels. There's a small twist in that players can re-use dice on a rolled 6, so there's a chance of dealing more than 1 strike per die. The following table shows the expected strikes per die value that you'd need to roll to succeed.

I was super surprised when I calculated this table! Somehow a die with 1/6 chance to get any particular value, when re-rolling the max value, gives you average strikes in increments of 1/5th! Neat!

With this table, you can now calculate the power of gaining extra strikes or an advantage (which effectively lowers the value needed for success by 1) for the rest of the turn and compare it to just rolling all of your dice normally. I ended up bounding the power of abilities to no more than 3 times as powerful as without them. That gave me an upper limit on what a single Talent could do.

Balance Isn't That Simple

There are two major flaws with this balancing system, sadly:

1. The fewer dice you have in your action pool, the less powerful spending dice to gain power for the rest of your turn is. Some talents mathematically aren't worth using when you just have your default 3 dice. I couldn't make them more powerful, though, because then they'd be over my power limit when you had 6 dice.
2. The strength of most talents is dependent on the die value needed for success. Almost always, talents are relatively more powerful against higher difficulty levels. Most talents are "balanced" against the max difficulty of 6, which means that they can often not be worth using when the difficulty is lower.

With so few possibilities on the dice, and with keeping the numbers low, there's not much I can do to address these flaws. I just kinda have to hope that talents don't feel particularly weak in normal circumstances. In the playtests I've done, my players haven't complained about talents feeling sub-optimal, which is nice. But I don't know if that would change over time.

Besides the flaws, there's one last interesting quirk about this dice system. The benefits needed to justify a talent's cost scales faster than linear. In other words, using a balanced 1-cost talent three times is likely to be weaker than using a 3-cost talent once. There's a built-in diminishing returns. Which makes sense if you think about it. A 5-cost buff to 1 action needs to match the power of a 4-cost buff to 2 actions. Double the actions but only 1 die more expensive.

I think this is great, actually. In TGitR, you can activate a talent multiple times and stack their effects as much as you want as long as you can pay the cost. Because stacking is inherently limited, I don't have to worry about making a talent that is balanced when used once but broken when stacked multiple times. It all kinda just works out! And I really do love the idea that you can stack the same ability multiple times. It's one less restrictive rule to write, one less exception for players to need to remember, and one less worry I as the designer have when trying to write new talents.