I had the idea of commissioning some custom versions of FFG's custom dice, for use at my table:

-I wished to change the games eight sided difficulty/ability into ten sided dice, with the same number and type of symbols as its pre-existing die type, with the exception of addi ng a number ranging from 1 to 10 on each face (so that these dice may also be used as a D10/D100, if needed).

-I also wished to alter the Proficiency/Challenge dice from their D12 form to a D20, again not adding any addition symbols on the new additional die faces aside from adding numbers ranging from 1-20 (so that should I ever conceive of a reason to use a D20* I may).

I wanted to get people’s opinions on how this might skew the odds/results of each die, or if I should add any additional symbols when I “bump up” the die type.

*I sometimes use a D20 for few custom charts a have crafted.

The problem with what you are suggestion would be having to reroll those "Blank" results should they land on a side that only shows a number. And how would you tell a blank roll (a result on a normal EotE dice that shows neither success/failure/threat/advantage/triumph/despair) from a result that was blank because you bumped up the size of the dice and it provided extra faces.

If you treat those sides as blank then you begin to mess with the probability that FFG has spent a lot of effort and time fine tuning.

I've poked around with the probabilities of different home brew dice ideas and I can say that tweaking the number of faces on a die can really shift how it scales when you stack multiple dice in a pool.

To use a really basic example, look at Triumph/Despair.

On the FFG d12, the probability of rolling Triumph is 8.33%. Putting that single Triumph on a d20 drops the probability to 5%. You might compensate by adding a second Triumph face, bumping the probability up to 10%. That gets you to within 2%, which might be a tolerable difference, but that gap widens as you add dice:

The chance for at least one Triumph on:

• FFG 4d12 is approx 29%
• 4d20 (with 1 Triumph face) is approx 19%
• 4d20 (with 2 Triumph faces) is approx 35%

And this is just for a single, uncancelled symbol. When you start throwing in dice whose symbols start cancelling each other out, the probability starts curving in any number of ways depending of the types and quantities of dice in the pool.

That's not to say you couldn't find a d20 configuration that keeps the game playable, you just may find that game gets easier (or harder) as players advance and more dice are added to the pool.

what? you don't want MORE dice?

A standard set of polyhedral dice numbers (d4, d10, d10 (tens), 2d6, d12, d20) can be as low as $5. also you are really nerfing or changing the odds by adding more faces. and the cost of custom dice just so you dont have to bring an extra of "Normal" dice seems rather silly

Thanks everyone for the feedback thus far.

what? you don't want MORE dice?

A standard set of polyhedral dice numbers (d4, d10, d10 (tens), 2d6, d12, d20) can be as low as $5. also you are really nerfing or changing the odds by adding more faces. and the cost of custom dice just so you dont have to bring an extra of "Normal" dice seems rather silly

I'll grant it is a wee bit silly if that where my only reasoning: but the small numbers I would add (basically as small exponents in a corner) are really just a small bonus that I felt like adding anyway if I where to go and increase the number of faces on any given die. I'm certainly not basing this on cost or dislike of FFGs current die odds (I own six sets or so of Star Wars dice, and will likely end up getting a few more). I wish to fiddle with the number of facings for use with homebrew rules/settings.

I've poked around with the probabilities of different home brew dice ideas and I can say that tweaking the number of faces on a die can really shift how it scales when you stack multiple dice in a pool.

To use a really basic example, look at Triumph/Despair.

On the FFG d12, the probability of rolling Triumph is 8.33%. Putting that single Triumph on a d20 drops the probability to 5%. You might compensate by adding a second Triumph face, bumping the probability up to 10%. That gets you to within 2%, which might be a tolerable difference, but that gap widens as you add dice:

The chance for at least one Triumph on:

• FFG 4d12 is approx 29%
• 4d20 (with 1 Triumph face) is approx 19%
• 4d20 (with 2 Triumph faces) is approx 35%

And this is just for a single, uncancelled symbol. When you start throwing in dice whose symbols start cancelling each other out, the probability starts curving in any number of ways depending of the types and quantities of dice in the pool.

That's not to say you couldn't find a d20 configuration that keeps the game playable, you just may find that game gets easier (or harder) as players advance and more dice are added to the pool.

As for the topic of I was sure that I would see a decrease in the odds of any particular result on each but as I was altering both the negative and positive dice uniformly, I figured it would all essentially come out in the wash.

If I where to add any new results to any particular dice I was thinking of adding an additional success and failure to the D20 versions of the proficiency and challenge dice respectively.

As for the topic of I was sure that I would see a decrease in the odds of any particular result on each but as I was altering both the negative and positive dice uniformly, I figured it would all essentially come out in the wash.

If I where to add any new results to any particular dice I was thinking of adding an additional success and failure to the D20 versions of the proficiency and challenge dice respectively.

The problem you'll encounter, though, is that you're increasing the chances of a "net 0" result, which is still a failure. You'll go from 1 Blank on a D12 to 8 Blanks on a D20, which increases the chance that you'll roll no successes at all.

If you're committed to this, I'd recommend filling out those empty spaces with a proportionate variety of symbols. You'll probably want to use something like anydice.com to run your designs through their paces before you commit to having the dice etched. Otherwise, your sessions will turn into either fail fests or bloodbaths depending on how the results skew.

Why not just roll a d20 when you need to use your d20 based chart?

I don't see any reason to have custom dice made. Just seems like it would be a lot hassle for little results. That's just me.

As I said before (although I'm not sure I articulated it clearly) the numbers that would be placed on the dice are not really the point of reworking the dice, more a bonus should it prove workable to include.

The problem you'll encounter, though, is that you're increasing the chances of a "net 0" result, which is still a failure. You'll go from 1 Blank on a D12 to 8 Blanks on a D20, which increases the chance that you'll roll no successes at all.

If you're committed to this, I'd recommend filling out those empty spaces with a proportionate variety of symbols. You'll probably want to use something like anydice.com to run your designs through their paces before you commit to having the dice etched. Otherwise, your sessions will turn into either fail fests or bloodbaths depending on how the results skew.

I do see your point, so with that in mind I may be looking at the following configurations pending a bit of trial and error:

Proficiency/Challenge:

2- Blanks (10%)

6- Success / Failure (30%)

3- Twin Successes / Twin Failures (15%)

2- Advantage/Threat (10%)

3-Success & Advantage/Failure & threat (15%)

3-Twin Advantages/Twin Threats (15%)

1-Triumph/Despair (5%)

Overall odds of generating either 'Success' or 'Failure' (not including Triumph/Despair): 60%

Overall odds of generating either 'Advantage' or 'Threat' (not including Triumph/Despair): 40%

Ability:

2- Blanks (20%)

3- Success (30%)

1- Twin Successes (10%)

2- Advantage (20%)

1- Success & Advantage (10%)

1- Twin Advantage (20%)

Overall odds of generating 'Success': 50%

Overall odds of generating 'Advantage':50%

Difficulty:

2- Blanks (20%)

2- Failure (20%)

1- Twin Failures (10%)

3- Threat (30%)

1- Failure & Threat (10%)

1- Twin Threats (10%)

Overall odds of generating 'Failure': 40%

Overall odds of generating 'Threat': 50%

I'm still fiddling with these configurations, and still not quite happy with them, It was (and to a degree still is) my philosophy that by adding a bunch of blanks, yes I've lowered the odds on the "good" dice, but at the same time lowered the odds on the 'bad" dice by the same amount kinda cancelling out the effect.

What sparked the idea for increasing the Die type's number faces was an idea that by lowing the odds of the dice giving any particular result at all I might be able to introduce attributes over six in play without seeing the dice system breakdown when more then six of each type (good & bad) see themselves rolled in a pool.

I'm still fiddling with these configurations, and still not quite happy with them, It was (and to a degree still is) my philosophy that by adding a bunch of blanks, yes I've lowered the odds on the "good" dice, but at the same time lowered the odds on the 'bad" dice by the same amount kinda cancelling out the effect.

What sparked the idea for increasing the Die type's number faces was an idea that by lowing the odds of the dice giving any particular result at all I might be able to introduce attributes over six in play without seeing the dice system breakdown when more then six of each type (good & bad) see themselves rolled in a pool.

That's ... interesting ...

I think what it would do is just stretch out the progression curve, so that:

• New Attribute 10 is equivalent to FFG Attribute 6
• New 8 is equivalent to FFG 5
• New 6 = FFG 4
• Etc.

Rank 3 would have to be your new "baseline human" with 2's being "below average" and 1's being "why bother rolling?" I imagine you'd be altering the XP rules to allow players to raise their attributes accordingly.

The other effect would be that, with more dice in play, it raises the potential for more net successes. So, even though your "New Rank" 8 gunslinger has the success/fail rate of an "FFG Rank" 5, he has the possibility of scoring up to 16 successes, rather than 10.

Like I said, it's interesting.

Without digging too deep into the numbers, I wouldn't know if there are any "gotchas" lurking in the stats, but if your intent is to stretch out the attributes like that, you might be on the right track with your original plan. Like a few others have said, my only other question would be "why?" but then again, you probably have wider home brew ideas where it all makes sense.

Hope they work out.

I love FFG's system, but for some reason I always feel compelled to tinker with games even when I am otherwise satisfied with the game. My idea is to create a more generic system for my own use, I like the idea of a more granular scale of one to ten. The One to Ten scale is much more granular creating more room for upgrades like extensive Cyberwear/Biowear, the application racial attribute caps, appropriately stated house Cats and more.

I am a bit of an RPG magpie so I like to hold onto mechanics that I like. My Homebrew System's overall Outline would make use of the modified version of FFG's 'Narrative Dice' I am trying to develop here, the Attributes from AEG's 'Legend of the Five Rings', and the skill advancement mechanic from Luke Crane's 'Burning Wheel'.

Back to the threads core issue: I am seeing a problem develop in my tests of my altered the die types: upgrading from Ability/Difficulty into Proficiency/Challenge becomes a trap as you are less likely to get an outcome from either as apposed to the "lesser" dice. My current thoughts to correct this probability issue are to attempt a reworking of the dice a small degree to give each die type a 40% of rolling a blank.