D&D General Blending individual checks into group checks

[Li Shenron]

This is a small idea that came to my mind, although I haven't needed to use it yet.

It's based on my general fandom of group checks to resolve several exploration tasks, and on the potential worry of players "rounding up" after one of them attempts something and fails.

Consider the iconic case of a stuck door: the strongest PC tries to open it, the DM has decided that the result should be random, and calls for an ability check. The strongest PC fail, so another player has the idea of their PC try but fails as well... at this point the DM has already opened the Pandora box of letting others try, and sure enough every other PC is now going to, largely increasing the odds.

The problem here is that what matters to me as a DM is the probability (not a precise value but at least a ballpark). If I grant the first PC a check, I set the DC so that there is a certain probability of success. Letting the second PC try is roughly similar to granting advantage, as in allowing the Working Together option (which is DM's call, not player's entitlement). This is already generous but let's say that if I would have allowed Working Together, then I am also probably ok with a second PC try. But this kinds of set the precedent, and how am I now going to justify to a third or fourth PC that I don't allow them? If I do allow them however, my probability ballpark is gone, success almost guaranteed, and why did I even ask a check to the first PC? BUT sometimes I DO want some outcomes to be random and not predetermined by me.

So here came my small idea...

What if, after the second PC either trying on their own or granting advantage to the first with Working Together, the resolution blends into a group check? It means, all the checks now count together and half of them need to succeed. The third PC trying won't be able to succeed alone because the first 2 already failed, so both the 3rd and a 4th need to succeed. If only one of them does, an additional 5th and 6th would be needed. This kinds of smooths the probability out again.

 

[Charlaquin]

This is a small idea that came to my mind, although I haven't needed to use it yet.

It's based on my general fandom of group checks to resolve several exploration tasks, and on the potential worry of players "rounding up" after one of them attempts something and fails.

Consider the iconic case of a stuck door: the strongest PC tries to open it, the DM has decided that the result should be random, and calls for an ability check. The strongest PC fail, so another player has the idea of their PC try but fails as well... at this point the DM has already opened the Pandora box of letting others try, and sure enough every other PC is now going to, largely increasing the odds.

The problem here is that what matters to me as a DM is the probability (not a precise value but at least a ballpark). If I grant the first PC a check, I set the DC so that there is a certain probability of success. Letting the second PC try is roughly similar to granting advantage, as in allowing the Working Together option (which is DM's call, not player's entitlement). This is already generous but let's say that if I would have allowed Working Together, then I am also probably ok with a second PC try. But this kinds of set the precedent, and how am I now going to justify to a third or fourth PC that I don't allow them? If I do allow them however, my probability ballpark is gone, success almost guaranteed, and why did I even ask a check to the first PC? BUT sometimes I DO want some outcomes to be random and not predetermined by me.

So here came my small idea...

What if, after the second PC either trying on their own or granting advantage to the first with Working Together, the resolution blends into a group check? It means, all the checks now count together and half of them need to succeed. The third PC trying won't be able to succeed alone because the first 2 already failed, so both the 3rd and a 4th need to succeed. If only one of them does, an additional 5th and 6th would be needed. This kinds of smooths the probability out again.

Yep, this is also how I’ve started handling “do I know anything about [subject]?” checks. If one player makes the check, fine. If a second player asks if they can try too, they can make the same check, and together they sort out the correct information if either succeeds, but neither can remember it if they both fail. This is mathematically identical to advantage if both have the same bonus to the check, or slightly worse if one has a lower bonus. If any more players ask if they can try too, they can, but unless at least half of them succeed, there’s too much conflicting information being contributed for the group to confidently sort out the truth from the misinformation.

The mathematical effect of this, by the way, is that it’s always best to have the smallest even number of people possible working together on any given task. Adding an extra contributor only hurts the group’s chances overall if the total number of contributors is odd, and if the total number is even, the group’s chances are better than they would be with one fewer, and worse than they would be with two fewer. This is the case for all combinations of DCs and bonuses (unless the DC is impossible for one or more contributors to succeed at, obviously).

 

[aco175]

In practice, the PC with the highest STR already tried to open the door, maybe using Athletics so the door should be pretty hard to open if he failed. Would it be easier to raise the DC check for each person afterwards. Would this offset the math of more people eventually rolling something high if the DC increases by 1 or 2 each time?

 

[TheSword]

The problem is that the swing of a +1 to +20 modifier depending on the dice roll is just too great to allow repeated re-rolls. While artificially adjusting DCs feels arbitrary.

I would allow someone else to try but only if they are more skilled than the first person to try.

Of course the real solution is to put consequences to failure but that can require a great deal of work.

 

[Charlaquin]

The problem is that the swing of a +1 to +20 modifier depending on the dice roll is just too great to allow repeated re-rolls. While artificially adjusting DCs feels arbitrary.

I would allow someone else to try but only if they are more skilled than the first person to try.

Of course the real solution is to put consequences to failure but that can require a great deal of work.

The thread wasn’t about rerolls, it was about using group checks to resolve when multiple characters attempt the same action.

 

[Lakesidefantasy]

This is what I call "Walking the Chimaera", because, as a Dungeon Master, it feels like the party is one body with many heads.

I would suggest noise, loud noise that attracts monsters for each failed attempt. If that's not going to help the situation then just allow them to beach the door. I mean, what's stopping them from just chopping a hole through the door?

 

[AK81]

I like the idea of it turning into a group roll. That's a good solution and I even think it supports the narrative. If the strongest couldn't do it by themselves, then they need more people working together. So a group roll fits perfectly.

 

[Lakesidefantasy]

I like the blended group check. Maybe after the second person tries and fails you declare that this is now a group check and we have two failures.

 

[aco175]

I like the blended group check. Maybe after the second person tries and fails you declare that this is now a group check and we have two failures.

Would you allow the PCs that failed to be part of the group check? I'm thinking yes since he may have failed on his own, but can add something as part of the group. I would still count it as a failure. Not sure if I would group check after one or two people try.

 

[Lakesidefantasy]

Would you allow the PCs that failed to be part of the group check? I'm thinking yes since he may have failed on his own, but can add something as part of the group. I would still count it as a failure. Not sure if I would group check after one or two people try.

That would not be blended.