A d20 alternate... the "Double Dozen", but asking for help.

[Dirty n Evil]

First off, I'd like to introduce myself. I am the unluckiest RPG player ever. I break the laws of probability with how often I roll a natural "1" in gaming sessions. There is no such thing as a "gimme" situation for me, because I will invariably roll a "1" at the worst possible time - every time. Action dice aren't enough to off set what an unlucky dice roller I am.

As you can imagine, experience has made me quite calloused against the arbitrary d20, with its equal chance of supreme successes or utter failures. Like many, our group has joked about the low usage of the d12, and I once jokingly commented that I would solve that problem and create a game system based off the d12! Off of that cast away comment, an idea started to form.

I plan to make a minor change in my upcoming campaign in which the d20 isn't rolled, but instead 2d12. Or, as I've been affectionately thinking of it, the "Double Dozen" system. It feels fairly straightforwards, to change DC's from a 10 base to a 12 base... yes, mathematically this still gives a slight preference towards successes but the difference is small enough that I'm comfortable with it. I like how this alternate system creates a very subtle bell curve of odds, with more rolls falling into the average than on the average d20.

The one aspect of this new system that I'm having hesitations about are weapon criticals. I want a system that works smoothly and quickly, and I haven't come up with a solution for this just yet. My first idea was clumsy... that criticals would move downwards from the highest possible rolls dependant on their range. A 20 would require two 12's, a 19 would require at the least a 12 and an 11, an 18 would require at least two 11's, ect. I could see this slowing down gameplay, the last thing a new system wants.

I've privately also wanted a system that didn't require seperate dice to figure out damage - why doesn't a high roll to hit (seemingly accurate) do more damage? I've considered a system where one damage was inflicted for every point of result over the target AC, up to a maximum the weapon could inflict. The problem I saw as that high AC targets would be very, very slow to whittle down. Perhaps if the bonus to damage from Strength or magical plusses was added not to the maximum potential damage, but added to the damage calculated after the rolled result? But that still doesn't address criticals. :(

Any constructive thoughts or suggestions are welcome... thanks in advance!

[Milskidasith]

I'd suggest you use the 3d6 variant proposed in the online SRD and Unearthed Arcana.

[Yakk]

So 2d12 differs from 1d20 by more than just a +3 shift. It also has a higher central tendency.

If your system revolves around 50-50 chances, you should take this into account. Lets look at how large it is!

Variance(1d20) = (20^2-1)/12 = 399/12 = 33.25
Variance(2d12) = 2*Variance(1d12) = 2 * (12^2-1)/12 = 143/6 =~ 23.83

Standard Deviation(1d20) =~ 5.77
Standard Deviation(2d12) =~ 4.88

SD(d20)/SD(2d12) =~ 1.18

An 18% difference in the "spread size" is probably small enough to ignore -- it is roughly the difference between a +5 modifier and a +6 modifier (or a +6 and a +7).

---

Next, for your critical problem. You have two dice you are rolling, so you might as well us that: I'd go with doubles.

---

Using your to-hit roll for your damage roll is problematic, because the time it takes to do the math (and make sure it is accurate) tends to take longer than just rolling another set of dice. With "0-1" hit/miss systems, if you roll really good and you aren't quite done adding your modifiers, but you hit by a large amount, you can stop and go onto damage resolution -- in systems where every point counts on your to-hit roll, you have to calculate it exactly.

What you could do is something along the lines of "your damage is based on one of the two dice you rolled to-hit". A counter-intuitive system would have low rolls being better for damage: then an easy swing would more likely contain a low roll, which ups your average damage on a hit. This is often avoided because people like having a simple rule of "bigger=better".

Another approach could be a mixture of a die-pool and sum-over system. Imagine if your attack was a certain number of d12s.

To hit you have to pass a target number.

You roll 2d12 to start. Then you can pick one up and roll another d12 at the cost of one of your attack dice. You repeat until your blow lands.

Any remaining attack dice can are then converted to damage dice.

Downside: places decision points all over the place.

Upside: is pretty continuous in how it converts accuracy to damage.

[Dirty n Evil]

Thank you for the input, it's appreciated. I'll consider your suggestions, however this one point didn't quite capture the heart of what I was pondering.


Next, for your critical problem. You have two dice you are rolling, so you might as well us that: I'd go with doubles.

This would work fine if all weapons had the same threat range for their criticals. Not to mention how it makes Imp. Critical useless. As you can see, the criticals are the trickiest point in this conversion.

[Draz74]

I'd suggest you use the 3d6 variant proposed in the online SRD and Unearthed Arcana.

Too different from the normal d20, both in distribution (too bell-curved for my taste) and in range (3-18 is a lot different from 1-20).

I recommend d8+d12. For one thing, you can still claim you came up with a d12-based system. But figuring out e.g. critical threat ranges might be a little easier than with 2d12, since the maximum roll (20) is still the same.

[Dirty n Evil]

I haven't worked through the math of it yet, but I did just have an idea on how to solve the criticals problem...

With the DC's that normally were 10, I said to just add 2 and make them 12's. Which is a nice consistant number, considering that one is using the d12 as the primary dice. But what if the criticals solution was as easy to solve? Before, I was thinking too specifically about the number rolled on the d12's. A 20 threat range requiring a 12, a 20-19 requiring at least one 12 and either one 12 or an 11, etc. Then, I realized perhaps I was looking at things with a bit of myopia.

What if it wasn't the specific number on the d12's rolled... but rather their totalled result? I allude to this above with the DC's, about how all I did was add 2 to the DC's. What if instead of requiring to roll a 20, they had to roll a combined total of 22 between the two dice? That would be easiest enough to figure out, since one is already adding the dice up to calculate whether they hit or not. However, I think the percentage times a result of 22 or greater will be rolled is too frequent. I'm going to do the math on what a 23 might look like, and how it might look for threat ranges if they use a 23 as the base instead of a natural 20. Such that a 20=23, 19=22, etc. It clearly makes a weapon with a greater critical range even more dangerous right away.

[Milskidasith]

I haven't worked through the math of it yet, but I did just have an idea on how to solve the criticals problem...

With the DC's that normally were 10, I said to just add 2 and make them 12's. Which is a nice consistant number, considering that one is using the d12 as the primary dice. But what if the criticals solution was as easy to solve? Before, I was thinking too specifically about the number rolled on the d12's. A 20 threat range requiring a 12, a 20-19 requiring at least one 12 and either one 12 or an 11, etc. Then, I realized perhaps I was looking at things with a bit of myopia.

What if it wasn't the specific number on the d12's rolled... but rather their totalled result? I allude to this above with the DC's, about how all I did was add 2 to the DC's. What if instead of requiring to roll a 20, they had to roll a combined total of 22 between the two dice? That would be easiest enough to figure out, since one is already adding the dice up to calculate whether they hit or not. However, I think the percentage times a result of 22 or greater will be rolled is too frequent. I'm going to do the math on what a 23 might look like, and how it might look for threat ranges if they use a 23 as the base instead of a natural 20. Such that a 20=23, 19=22, etc. It clearly makes a weapon with a greater critical range even more dangerous right away.

Actually, that makes a weapon with a greater critical range less dangerous... it's a 1/36 chance to get a 22 with a 19-20 weapon, which means it's less likely to crit than a normal 20-20 crit range weapon.

Again, I'd just recommend the 3d6 variant. It already does everything you want it to do while still keeping the crits the same and the chance of getting a success (by rolling a 10) even instead of adding 2 to all DCs.

Adding 2 to all DCs is also fairly hard when you factor in other variants, like "players make all rolls." Even without that, it's still clunky. 3d6 is much more elegant.