Rolling two dice and taking the highest (2DTH from here on) is one of my favorite recent discoveries. I never saw this back in my Second Edition days. The first place I came across it was probably at Grognardia, perhaps one of the Dwimmermount session reports. I would be interested in knowing the ultimate provenance though. Is 2DTH a recent innovation, or can it be found in any of the classic writings? Philotomy has a similar rule in his OD&D musings, so maybe that is where it came from.
I like 2DTH so much because it allows you to skew probabilities in a particular direction without introducing bonuses or eliminating any possibilities. Also, though you are literally rolling more dice, it still feels like a single roll, and thus does not seem to bog pacing down as something like an additional attack roll might. Other editions have addressed similar problems by adding bonuses (the Third Edition family and Second Edition to a slightly lesser degree) or eliminating randomness altogether (such as HP in Fourth Edition).
Places where I think this mechanic is appropriate:
- Damage: two-handed weapons or dual-wielding
- First level HP
I’m sure there are many more as well.
The d4 is the simplest interesting case (the curve shift is not as obvious with a d2).
This gives the following probabilities:
- 7 in 16 chance of a 4 (~ 44%)
- 5 in 16 chance of a 3 (~ 31%)
- 3 in 16 chance of a 2 (~ 19%)
- 1 in 16 chance of a 1 (~ 6%)
Obviously, two dice take lowest (2DTL) can also be used to skew the probability curve in the other direction, though I have found less use for this. Perhaps for situational modifiers.