D&D 5E How fantastic are natural 1's?

[R_J_K75]

On a side note, I do sometimes add extra color to failures for skill checks if someone fails the DC by 10 or more. So if the party is sneaking, someone knocks over a metal cup which causes a pot to tip over which causes a loud cascade of a dozen pots, pans and glassware to fall to the floor.

But it doesn't matter if they missed by that much because they rolled a 1 or a 5.

I tend to use that as a gage of failure too. Truly that is what the skill system is built on. A lot of times I'll use the fail forward where the player passes the desired skill check but with some unintended consequences.

 

[Ovinomancer]

Not sure how you get 37% on 9 attacks. There's a 5% chance on each and every swing of a fumble, unless you're assuming the fumble stops every subsequent attack? Which ... I guess it might. Also makes it an even harsher penalty.

In any case, like I said, I did try a variation of your confirmation rule and it worked okay but still didn't address all of the issues. Typically ranged attackers that roll a 1 hit an ally. Frequently they automatically hit an ally no matter what that ally's AC is, which honestly is penalizing the ally not the archer.

I guess I simply have different ways of adding excitement to the game, if there's only a 1 in 144 chance of failure at what point does it become effectively no chance and not worth rolling?

Because the probability is not simply additive. The easy way to determine odds of rolling at least 1 one in a 9 attack routine is to figure the odds of hittimg all 9 times and subtract that from 1. Those odds are .95^9= 0.63. Subtract from 1 and you get a 37% chance of rolling at least 1 one.

 

[DND_Reborn]

Not sure how you get 37% on 9 attacks.

It's a binomial distribution. Simply take 100% minus the chance of getting zero natural 1. Or 1 - 0.95^9 = 1 - 0.6302 == 0.37 (ish) or 37%.

With Excel it is "=1-BINOM.DIST(0,9,0.05,FALSE)", which is 0 success, 9 trials, 0.05 probability of success, an not cumulative.

How are you getting up to 45%?

EDIT: LOL sniped by @Ovinomancer !

 

[Oofta]

I think the problem I have with the basic premise (other than the fact that I hate automatic fumbles on 1's obviously) is that I try to run a very descriptive game. It doesn't matter if the PC succeeds incredibly well, or if the PC fails by more than 10.

For example in my game yesterday the rogue found a glyph of warding trap on a door they needed to open. I let the wizard help her with her disarm attempt and she got a 29 on her check to disable the trap. I decided that not only had she disabled the glyph, but no one could tell that it had been disabled.

The example of the cup causing a cascade of crashing plates happened a couple of games ago as the group was trying to sneak into a tavern through the back door. It wasn't a high DC and even failing up to a certain point would only have raised suspicion of the sleeping dog.

So fun descriptions of extreme failure (and success) happen all the time. Either I'll come up with something or the player will so they don't particularly stand out.

 

[Crimson Longinus]

Because the probability is not simply additive. The easy way to determine odds of rolling at least 1 one in a 9 attack routine is to figure the odds of hittimg all 9 times and subtract that from 1. Those odds are .95^9= 0.63. Subtract from 1 and you get a 37% chance of rolling at least 1 one.

Whilst true, Oofta's point stands: that fumbles disproportionally penalise characters with multiple attacks and make high level fighters total jokes.

 

[DND_Reborn]

Typically ranged attackers that roll a 1 hit an ally. Frequently they automatically hit an ally no matter what that ally's AC is, which honestly is penalizing the ally not the archer.

That would be a "disaster" to hit an ally. A mishap is you drop your bow, your quiver tips over and some arrows fall out, etc.

I guess I simply have different ways of adding excitement to the game, if there's only a 1 in 144 chance of failure at what point does it become effectively no chance and not worth rolling?

And honestly, this is the point I struggle with in much of the game, not just this house-rule. At very high levels, many ACs become nearly automatic hits and it is boring. We've implemented A LOT of house-rules to make the game more exciting again and not as easy/automatic. shrug

 

[Oofta]

It's a binomial distribution. Simply take 100% minus the chance of getting zero natural 1. Or 1 - 0.95^9 = 1 - 0.6302 == 0.37 (ish) or 37%.

With Excel it is "=1-BINOM.DIST(0,9,0.05,FALSE)", which is 0 success, 9 trials, 0.05 probability of success, an not cumulative.

How are you getting up to 45%?

EDIT: LOL sniped by @Ovinomancer !

I don't claim to be a statistician. I don't even play one on TV. I still don't follow because a 1 always misses and there is a 5% chance of a 1 every time you roll a D20. If you flip a coin it doesn't matter if it was heads or tails the previous flip, it's still a 50% chance of heads or tails. But it's lies, damn lies and statistics so ... sure. Still more than I would care for.

 

[Ovinomancer]

Whilst true, Oofta's point stands: that fumbles disproportionally penalise characters with multiple attacks and make high level fighters total jokes.

Okay. Did I say it didn't?

 

[Crimson Longinus]

Okay. Did I say it didn't?

No, it just seems to me that that quibbling about the exact percentage is a tad besides the point.

 

[DND_Reborn]

I think the problem I have with the basic premise (other than the fact that I hate automatic fumbles on 1's obviously) is that I try to run a very descriptive game. It doesn't matter if the PC succeeds incredibly well, or if the PC fails by more than 10.

So, (and maybe you said this already) do you use critical hits? It is the same logic. It doesn't matter on a d20 roll if you make the target number (AC or DC) by exactly the number or by 5 or by the maximum.

I don't claim to be a statistician. I don't even play one on TV. I still don't follow because a 1 always misses and there is a 5% chance of a 1 every time you roll a D20. If you flip a coin it doesn't matter if it was heads or tails the previous flip, it's still a 50% chance of heads or tails. But it's lies, damn lies and statistics so ... sure. Still more than I would care for.

Sorry, I am and when other people throw out numbers that aren't accurate I question their methods. My apologies.

Without the confirmation roll, I agree, it would be way too high. Without it, a fighter with 3 attacks would fumble nearly 15% of the time while a single attack only 5%. But, with the confirmation roll, it works out well.

As you said before, however, it raises the question of whether the additional rules/rollings make the game more fun or not? Since the base chance is only 1 in 20, it doesn't come up often, but when it does it adds extra tension I've always enjoyed and have always used in one form or another.

Whilst true, Oofta's point stands: that fumbles disproportionally penalise characters with multiple attacks and make high level fighters total jokes.

Not if confirmation rolls are used, but again then it becomes a question of whether the rules add to the fun or not. For many posters, obviously not. For me, yes.