Best way to make 4d6 open-ended?

[SwordofLux]

Why not just reroll matching numbers? Granted, extremely low results can become moderate results, but it's much simpler. If you match two, then reroll those two, if three, then three, etc. Also, it changes the bell curve a bit, but sometimes compromizes must be made in the name of simplicity. Personally, having to keep track of which colors work together to trigger a reroll is an added layer of complexity for players. I don't mind complicated mechanics, as long as they're smooth - that is, time isn't taken up determining the results of a die roll. I'd like to be able to tell at a glance the result, and then be able to immediately reroll exploding dice.
MJK

[Mike Holmes]

Hi MJK, welcome.

I assume you mean roll matches and add? How do you get open-ended low, then?

Mike

[SwordofLux]

Thanks for the welcome. And D'oh! I didn't realize the goal was both high and low open-ended. I suppose in that case, that any matches of 4 or 5 (4/4; 5/5; etc.) would reroll, and add, and any matches of 0 or 1 would reroll and subtract. For example, a roll of 3, 5, 1, 1 would result in a total of 10, and a reroll of two yielding say, 2, 4 giving a final result of 4 (10-6). This may or may not work depending on the stat bonuses. If I recall, the average stat was 10, and the average roll 10. So, with the lowest stat probably being around 5, and an extremely low roll being about 4, an average of 5 for 2D0-5, I think this could work.

Of course, it is possible to go negative, but not likely. (With a stat of 5, one would have to get an initial total roll of about 3 (1,2,0,0) in order for even an extreme roll on one pair (9 or 10) to cause a negative.) Interestingly enough, an extreme initial roll (16+/4-) will cause a shift in the same direction (ie, even higher or lower). So, with a stat of 5, an initial total roll of 0 will cause likely cause a -5 total. This may be good or bad, depending upon your design goals.

One could treat negatives either as negative rolls, if the margin of success was the important factor, rather than the actual number of the roll. Or, simply cancel out negatives and treat it as a roll of zero. But, there's a possible system - take it as you will.
MJK