No, it's very simple—the issue is that it's just a little less simple than the designers thought. In every case you roll dice such that your expected result is approximately your step number. Improve your step number, your expected result goes up. Great, right?

Unfortunately, no one told the designers about variance. As a result, you can get nonintuitive effects like reducing your chance of success at a given test if you improve your step number by one—the simplest example is the same one Savage Worlds is plagued by due to exploding dice with a k=k+1 bug and multiple die sizes.

To illustrate: at Step 3 you roll 1d4; at Step 4, 1d6. All dice are exploding. At Step 3 your odds of rolling a 6 or better are (1/4)*(2/4)=3/16. By contrast, your odds on a 1d6 are 1/6, or 3/18—by going up a step, you lose out on this particular TN due to the different shapes of the probability curves around the respective areas.

Admittedly, the fact that the dice are summed after exploding does make exact by-hand probability estimation more complicated, but it's very amenable to automation.

~J

You are thinking of the ED Classic line. The 3rd edition books are definitely not HUGE. (IMG:style_emoticons/default/wink.gif)

And both ED Classic and ED 3rd Edition are released and developed by RedBrick.

The 2 Windlings in our party would completely disagree with you on that point I would think...



I think the 3rd Edition is MUCH better than previous editions... It runs smoothly, and Disciplines are A lot More customizable than they ever were before...

As for Acrobatic Dance, I think it is quite useful myself...

And as for the Karma Dice... Have not really seen a big difference in the play style... those with lots of Karma use it, and those with less, save it for the important rolls...

Anyways...

You are indeed confusing them with the Red Brick Editions...
There are 4 Main Books for 3rd Edition... 2 for Players and 2 for GM's...

My group and I are sticking with EDC as our flavor of ED. Navigation was easily solved through sticky tabs on each chapter. For us it stayed close enough to ED1 while solving some of the problems. ED3 didn't do enough to clean up some of the issues with spells so the only thing I've borrowed from it was the modified chart on Journey to Life so our Wizard can't bring Elianar Messias and King Tav Korelsed back from the dead.

As to the math, the developers released their graphs and a probability by step vs. TN chart at the site at Redbrick. So you can see exactly what your chances are. ED3 eliminated the d4 and d20 from the system for the reasons that Kage pointed out. My table likes that wonkiness (but we also liked the SR3 TN 7 problem so YMMV).

Just getting rid of those two dice doesn't help—the problem exists anywhere you have exploding dice with an n=n+1 problem and two dice of sides k and k+2 at TN k+2, so you've still got it with the d6/d8, d8/d10, and d10/d12 pairs (there may be another issue with variance on the d20 that was fixed, but since there's no d18 it doesn't have the same issue).

~J

Then what were they trying to help by getting rid of the d4 and d20? I could spend the time to put together the relevant PDFs (not the file type) but it seems like you might have already done. So what did the removal of those two die-types do?



Edit: I very nearly bought all four of the new 3rd edition corebooks last Friday but restrained myself at the Checkout. For now, they languish in my Wish List.

I have no idea—all I know is that the dk/dk+2 with k=k+1 at TN k+2 issue (there really has to be a shorter way to describe that problem) is almost completely unaffected by the change, so either they didn't know what they were doing or they were trying to fix something different. I have no insight into which of these are true or what the "something different" might be.

~J

The d4 is so small that the problem mentioned above happens all the time. The problem with the d20 is that, unlike d6 -> d8 > d10 -> d12 -> 2d6, there's no easy way to slide the d20 in.

Not to mention that the variance between d12 d20 and 2d6 ment that almost always you would rather have 2d6 or 2d8 than a d20

Unless of course you tended to roll well, and then the yatzhee factor in the d20 was huge.

I am noticing the issue in a deadlands game I am playing in. My mad scientist recently got his mad science up to 5d12 but I usually get around a 10 result.(then after drawing cards ends up being like a nothing result.) I really don't know the rules of it but I notice I do about as well with my fighting of 4d6 since at least one d6 explodes almost every time. Though last game I got 18+ a lot on my mad science, not sure why.