I was responding to someone who had asked what I thought of Ironsworn’s dice mechanics, which compares a d6+modifier versus two d10s to determine degrees of success. The resulting distribution is similar to PbtA-style 2d6, so my response was to inquire whether the d10s had any other function in the system. Otherwise, it seems more complicated than just rolling 2d6 versus a fixed range of results.
It weights the mods quite differently. A +5 on the d6+M vs d10 & d10 has a chance of no hits (roll ⚀+5=5; 10s can still cause fail/weak/strong hit results), while the 2d6+5 vs 6-/7-10/11+ is a guaranteed weak or strong hit.
the 2d10 makes modifiers up to +9 viable for all three results, while +4 is the maximum capable of all three results on PBTA standard.
(I'd pondered the why myself.)
Cranking the examples:
M modifier [miss, weak, strong]
M= 0 [355, 190, 55][59.166%, 31.666%, 9.166%]
M= 1 [271, 238, 91][45.166%, 39.666%, 15.166%]
M= 2 [199, 262, 139][33.166%, 43.666%, 23.166%]
M= 3 [139, 262, 199][23.166%, 43.6665, 33.166%]
M= 4 [91, 238, 271][15.166%, 39.666%, 45.166%]
M= 5 [56, 208, 336][9.333%, 34.666%, 56.0%]
M= 6 [32, 176, 392][5.333%, 29.333%, 65.333%]
M= 7 [17, 146, 437][2.833%, 24.333%, 72.833%]
M= 8 [9, 122, 469][1.5%, 20.333%, 78.166%]
M= 9 [6, 108, 486][1.0%, 18.0%, 81.0%]
Note also: I've ignored the 10% chance of a twist; a twist will always be a failure or a strong hit, since it requires the d10's to match.
Now, similar for PBTA 2d6+M vs 7+/10+, of 36
M= 0 [15, 15, 6][41.666%, 41.666%, 16.666%]
M= 1 [10, 16, 10][27.777%, 44.444%, 27.777%]
M= 2 [6, 15, 15][16.666%, 41.666%, 41.666%]
M= 3 [3, 12, 21][8.333%, 33.333%, 58.333%]
M= 4 [1, 9, 26][2.777%, 25%, 72.222%]
M= 5 [0, 6, 30][0.0%, 16.666%, 83.333%]
M= 6 [0, 3, 33][0.0%, 8.333%, 91.666%]
M= 7 [0, 1, 35][0.0%, 2.777%, 97.222%]
M= 8 [0, 0, 36][0.0%, 0.0%, 100%]
M= 9 [0, 0, 36][0.0%, 0.0%, 100%]
(All stats by running the total permutations in python.)