Help with the dice-roller (simulate specific rule) or statistics

Hellooooo everyone !! I was wondering if there could be anyone of help with a small obstacle that came up while I wanted to use a new game system for a campaign soon to be started. Here's the deal : ----------------- The system is based on Xd6. Each player rolls a variable number of d6 depending on the abilities of the PC (from 1 to 10, usually around 5 or 6 wich is the average). For each result of 1, 2 or 3, there is no (0) success. For each result of 4 or 5, there is one (1) success. For each result of 6, there is a double-success (2 successes). Roll example : -------------- A character has a Dexterity ability at 4 and a Climbing skill at 3, so he should roll 7d6 to attempt overcome a wall blocking his path. The results are the following : 1, 4, 6, 2, 3, 3, 6 If I use the basic roll formula for successes equal or higher than 4, I will end up with a result of 3 successes : /r 7d6>4 But the real result according to the game system should be 5 successes because of the 6s that are counted double : ??? I have tried to simulate this with a number of different possibilities, but failed so far. Apparently, it may be impossible to reuse a roll with a different set of operators, so something along {[roll Xd6] then [compare each result of the roll to find 4s or higher and count them as 1 success] and then add [compare each result of the roll to find 6s and count them as 1 success]} is not going to work. I also tried bypassing the fact that the parser wants to add the results of the dice if you don't specifiy that it's a "successes" roll by doing this: /r 7d6>0 so it still gives me each individual results, but I failed to understand how I can reuse a line of the chat inside a future roll. my guess right now is that's it's not possible either. Finally, one of my friends/player suggested to roll two separate sets of Xd, one with the value of success at 4 or higher, the other with the value of success at 6. It "works", but to be honest I'm not sure if the probabilities are messed up or not, and neither of us are statisticians, or good at math in general, or quite intelligent overall ^^ /r [[7d6>4]] + [[7d6>6]] Intuitively, I would say it's weird and skewed up, but I've learned statistics are often counter-intuitive so... If anyone can certify me that probabilities are the same, then that would be the simplest solution so far, and I'm more than willing to roll that way. Any help/advise would be welcome ! Anyway, cheers and have fun :)

The simplest solution is to just roll the xd6 and sum them up yourself, the way you would at a table in real life. ;) If you're creating a custom character sheet (requires a Pro sub) the custom roll parsing would do the trick, though. If you want to pursue that avenue, here are some websites that could help... Creating Custom Character Sheets Custom Roll Parsing Forum Post about Custom Roll Parsing

Or you could create a rollable table that just generates the number of successes.  It would only have 3 entries, 0 with a weight of 3, 1 with a weight of 2, and 2 with a weight of 1.  Then instead of rolling [[xd6]], you would roll [[xt[tablename]]]].

There is a simpler solution to the previous suggestions: Use failure points to do this (described on the Dice Reference wiki article). The roll is a bit clunky, but will generate -1, 0, or 1 success per die, so then you add the number of dice to the roll, gettinh 0, 1, or 2 successes per die. /roll ?{How Many Dice|2}d6>6f<3 + ?{How Many Dice} or [[?{How Many Dice|2}d6>6f<3 + ?{How Many Dice}]] If you want to explicitly declare each attribute, replace the How Many Dice query with attribute references like /roll [[@{selected|dexterity} + @{selected|climbing}]]d6>6f<3 + [[@{selected|dexterity} + @{selected|climbing}] But this is likely too much trouble.

Wouahou, yes thank you ! Three lightning-quick and good answers, that's awesome ! :) @Finderski : yes, my first reflex was to manually count the sixes to add them to the total already calculated by the formula ^^ I'm on the verge of getting a subscription, so as to be able to make custom sheets, so I'll probably give it a shot in the more-or-less near future. Thanks for the links ! :) @Kraynic : very good idea indeed, it didn't even cross my mind to do it with a table, that's a really nice solution :) @GiGs : ah yes, I thaught about using the "anti-success" thing, but couldn't wrap my head around how to do it. You can see here plainly my gigormous skills at math and logic... Thank you so much ! Cheers everyone and have even more fun !! ^^