Plane Sailing said:????? I'm afraid I don't understand how you derived this bit? (not doubting you, just don't understand it). In the example you give (two attacks, both hitting on a 17) I'm sure that the chance of -at least one- hitting is 36%. (I wonder if we are talking at slightly cross purposes? In order to get the "average number of attacks", are you effectively adding in the attack "twice" in the circumstance where both hit? Although that would apparently count for the "missing 4%" between our two expectations, I don't think that is right... but I guess I'm probably talking about the "probability at this point", while you are talking about the "effective average over a long period of time" ??
Anyhow, as I mention at the end of my post, the issue *isn't* average damage - which is rarely important to know because of the significance of all the other variables - but how likely you are to get at least one hit in (which as you note at one point is sometimes the critical issue - if the target is on his last legs, with only a couple of hit points left, for instance).
Yes, I guess we were. Both the table in S&F and I were, in fact, talking about average damage per round (= average number of hits per round * average damage per hit). I'd question whether this is 'rarely important to know', as most D&D combats (especially involving monks) are likely to drag on for at least a few rounds. For example, in the situation above where the monk needed a 17 to hit with his two flurry attacks, he would need a 15 to hit with his one normal attack. Presuming a <4th level monk with a 14 strength, he'll be doing 1d6 + 2 or an average of 5.5 damage per hit. This works out to an average of 1.75 per round without the flurry, or 2.2 per round when flurrying. This is a much more significant difference than the .36 chance of hitting at least once in a given round with the flurry vs. the .3 chance of hitting at least once with the single attack. Especially if the target has at least ~6 hit points, where you can't expect to take it down with a single hit ...
Your method of calculation is correct for determining the chance of 'at least one hit', which (as I mentioned above) is another interesting piece of information. But I'd question whether it's more important information than 'average total damage' in most cases.