One of the issues I've got with WHFRP is the ever-expanding dice pools, and I've been toying with how to address this. My current thinking is to keep things relatively simple, and rather than matching up different dice types, largely leave the decision down to individual players as follows:

1) No more than 2 Challenge Dice per pool - any more subtract from either Stance, Characteristic or Expertise dice (players' choice).

2) Fortune or Misfortune dice - unlimited numbers of 1 type in a pool, no more than 1 of the alternative type. Pairs of Fortune / Misfortune dice cancel out until the pool complies with this requirement.
E.g. 4 Fortune and 2 Misfortune becomes 3 Fortune, 1 Misfortune.
4F, 3M ->2F, 1M.
4F, 0M -> 4F.

I was also wondering about going further and allowing 3 Fortune to remove 1 Challenge or 3 Misfortune to remove 1 Characteristic, Stance or Expertise. But I think this runs the risk of bringing additional complexity and will slow conflicts down with the need for players to evaluate the best use of their dice cancelling options, which I want to avoid.

Before taking the long way on doing prfound modifications on the mechanics of the game, ask yourself if you like the game enough as to walk that path, or what do you like from the game that keeps you in. May be there are other games out there which you will also enjoy, ones which won't need that kind of heavy modifications and will allow you to focus on other aspects like scenario preparation which is more gratefull for both GMs and players.

Said that, from what you propose I would say you need to get yourself a probability calculator for the dice of Warhammer 3, there are many out there. Then start to look to those things again. A few of the things you say, like cancelling one challenge with one blue dice is really unbalanced (probabilistically speaking).

A better way (less painful for you and already balanced) to reduce dice pool sizes would be to get the SW dice and its dice mechanics and bring it without* any changes into the Warhammer 3 game

*you will need to get rid of fortune dice in characteristics and the fortune dice due to specialization.

Cheers,

Yepes

Yepesnopes said:" May be there are other games out there which you will also enjoy, ones which won't need that kind of heavy modifications and will allow you to focus on other aspects like scenario preparation which is more gratefull for both GMs and players. "

Heavy modification? For any pool up to Unlimited Char/Stance dice, 2 challenge dice, 2 fortune dice, 2 misfortune dice: no change. It's only the larger, more unwieldy pools that I'm planning to shrink. Reducing Fortune / Misfortune keeps things a bit more random, as the dice sets match and so the more fo them there are, the more the result will cluster towards the mean.

Said that, from what you propose I would say you need to get yourself a probability calculator for the dice of Warhammer 3

Got one. Made a spreadsheet a year or two ago. But I just threw this suggestion up there, you're right I should just check it out.

A better way (less painful for you and already balanced) to reduce dice pool sizes would be to get the SW dice and its dice mechanics and bring it without* any changes into the Warhammer 3 game

And where's the fun in that? :-)

I have several rules in our house rules, that deal with points in this thread. Link in my signature.

Ripple effects you might want to keep an eye out for with Phild's proposed house-rule in play:

• Spellcasters will be boosted more than non-casters. Chaos stars are extra bad for wizards, and thanks to quick-casting they see more challenge dice than most other characters. Wizards will be very happy to trade out a lowly blue to eliminate a purple.
• Cards that are "balanced" by bane- or chaos-star-triggered drawbacks will get a boost. This may require more vigilance from the GM to keep those cards in check. In other words, think twice before letting your player's choose and resolve their own banes.
• First Aid checks will be much stronger in Phild's system. They and any of the other non-standard rolls where the total number of successes dictates the strength of the result (instead of just aiming for a 2- or 3-success plateau) will be boosted more by your rule than rolls that follow the default action-card structure. Using the optional "Higher Lethality" rule from the GM's Toolkit will alleviate this disparity, but also increase damage in the late campaign.
• The intoxicated condition is mildly undermined. No big deal.

The math behind these ripple effects follows…

phild said:

While that's certainly how the bell curve of a handful of normal numbered dice works, it's not actually what happens when you add more fortune and misfortune dice. Instead of pushing the results towards the mean, to some extent they actually push away.

IMHO, the interactions of the white and black dice are one of the coolest things about Warhammer's mechanics. In most other games, shooting from high ground in the rain (or any other combination of relatively equal positive and negative modifiers) is always a boring +0 net, but in warhammer those two equal but competing modifiers have 7 possible results. +1/+0, +0/+1, +1/-1, +0/+0, -1/+1, +0/-1 and -1/+0 are all completely viable and meaningfully different results from 1 white and 1 black.

Paired fortune and misfortune dice actually increase the odds of the least likely outcomes showing up. I'll illustrate with some sample dice pools.

4 characteristic and 1 challenge:
1+ success: 72%
3+ success: 17%
2+ boons: 20%
2+ banes: 4%

4 characteristic and 1 challenge PLUS 4 Fortune and 4 Misfortune
1+ success: 66%
3+ success: 25%
2+ boons: 28%
2+ banes: 8%

Adding four each of white and black reduced the chance of the least interesting / most common result (simple success with no boons or banes). The odds of getting interesting results lines like triple-success, double-boon or double-bane all went up, which is pretty cool.

Let's look at the effects on a larger dice pool…

3 conservative, 2 characteristic, 3 expertise and 3 challenge:
1+ success: 84%
3+ success: 53%
5+ success: 20%
2+ boons: 63%
3+ boons: 41%
2+ banes: 3%
3+ banes: 1%


3 conservative, 2 characteristic, 3 expertise and 3 challenge PLUS 8 Fortune and 8 Misfortune:
1+ success: 75%
3+ success: 36%
5+ success: 3%
2+ boons: 59%
3+ boons: 41%
2+ banes: 7%
3+ banes: 3%

In this case, the "clustering towards the mean" does exist but only in regards to the odds of scoring very large numbers of successes. Again the odds of scoring boons and banes have actually gone up.

A larger portion of the actions in the game feature interesting effects for scoring a multiple boons or banes than for scoring 5 instead of 3 successes, so this larger pool will effectively provide more diverse results (unless you use the "Higher Lethality" optional rule from the GMs toolkit).

In your system that last dice pool would be converted to: 3 conservative, 1 characteristic, 3 expertise , 2 challenge, 1 Fortune and 1 Misfortune:

1+ success: 88%
3+ success: 58%
5+ success: 21%
2+ boons: 66%
3+ boons: 43%
2+ banes: 2%
3+ banes: 0.3%

Overall success rate is up, and the potential for a roll that succeeds but at a cost (lots of banes or chaos stars) is greatly diminished. Chaos Stars drop from 33% likelihood on 3 purple to 23% on 2 purple. To me that seems less dramatic… but YMMV. If that works for you and your group, great, that's all that matters.

You're quite right, of course, and it's very interesting to see how the results don't normalise anywhere near as much as I expected. That'll teach me not to do the arithmetic first. The reason is those 3 null results on each Misfortune die, meaning there is always a decent chance of getting an outlying result.

E.g. with 3F+1M, there is +1/0/-1 distribution of 37%/35%/10%, whereas on 2F that is 44%/44%/0%. A real difference.

Now I still think there is some mileage in reducing the number of dice, but one might have to be a bit more subtle: looking at some comparative data I've run, I think the golden rules are:

1. Don't reduce any die below 2
2. Don't remove more than 2 of any die

So, you could go from 5F/3M down to 4F/2M without changing the probability distribution too much, and you can drop from 8F/6M to 6F/4M again without changing too much. Any more than that and you cut out some very real probabilities of some big numbers coming in.

Given these golden rules, the simple conclusion is this: you can't reduce Fortune dice and end up with results close to probability spread. But maybe this is a Feature of these changes, not a Flaw. Perhaps this simply acknowledges that while Fortune (or Misfortune) can influence events, its impact is limited.

More thought needed! Fun with numbers :-)