Does the matched speed weapons really have any noticeable impact? I realize the potential for an extra flurried attack, but does it happen often enough to really even be a consideration? Whether or not you can actually get two decent weapons of the same speed is a relative non issue though, because you can do that easily.

Hmm, won't this assumption work against the fast weapons? Consider a situation where flurry is not triggered within the uptime. The next possible time a new flurry can be triggered is the next attack (white or yellow), the time until this next attack will in general be smaller for faster weapons, thus you will spend less time as unflurried with fast weapons. Or did I go wrong somewhere?

Not sure how to model the effect of unflurried periods though.

Take into account, that "average" stats of "use-trinkets" like [Bloodlust Brooch], used in calculations are in fact the worst case scenario. That means, its good only in static fights, which last for (trinket's cd * n) minutes. For the Brooch it's 72ap + 1/6 * 278ap = 118.33ap, where 1/6 is taken from: 20secs of use divided by 2mins cd.

The best case scenario is fight which lasts for (trinket's cd * n + trinket's use-ability's time), for Brooch that would be for example 2min20sec fight, where it's value is rather 72ap + 2/7 * 278ap = 151ap, where 2/7 is taken from: 40secs of uses (one time at 0:00, and 2nd time at 2:00) divided by 2min20sec of fight.

Also, trinket's with use ability have natural advantage over "proc" ones in non-static fights, where u want to pump up your dps in certain moments.

That altogether probably makes [Berserker's Call] the best trinket for dps warrior in game after 2.3.

I'd also be glad to see some way of including sharpening stones in calculations, since +12wep dmg is far from unsignificant, and it favorizes fast weapons too, which are, right now, discriminated by the spreadsheet.

No it shouldn't. Probability to renew it regardless if it's up or not is the same, perhaps my wording was a bit ill. In another way it does not give fast weapons a correct uptime though, but that part is incredibly hard to model and is what I used my simulator for. There is a startuptime for all (regardless of critchance and weapons speed). This will be shorter with faster weapons and higher critrate. Ie, the theoretical uptime is too positive, it assumes you have flurry on from the beginning and there is no pauses in dps. However these errors are not that big (I would guess in the vicinity of 1% or less) and other factors matter more, such as if you have to run out or boss goes away for 12+ seconds you will loose flurry, and you have to get a first crit again. Anyways, I think that will be overcomplicating things, and making a full dps simulator would be easier than incorporate that in to forumale. Boss fight, generally have a greater impact on flurry uptime (how many bosses do you stand still 10 minutes and just go full out from start to end?).

Long story short: it works against fast weapons marginally (for other reasons though). It's not worth to try to fix.

If you're talking about adamantite sharpening stones, those are already included in the spreadsheet. It adds 7 damage (12*.625 rounded down) to your off hand's average damage.

The only movie I know of is Curse's BT video which shows several shots of their fury warrior DPSing: Curse(MYM.WoW) : Black Temple The Movie

On a separate matter I and others have noticed that if you increase your stats to the point where you get an extra HS in your cycle, dps will show a huge increase even if it was just an insignificant amount of stats that were added (I've had a 20 dps increase from a 5 hit rating addition). Perhaps a better way to model this would be using partial numbers in the cycle instead of whole numbers for excess rage. I'm not sure what is done with excess rage that isn't used in the cycle so I may be missing something.

After 2.3 which is best: fast offhands or slower ones?

Does the damage from having a slow offhand offset the benefits of having a faster one?

What about flurry up time. Does a faster offhand translate into an overall higher up time for flurry?

Why should it?
It's exactly the other way round as faster offhands consume instant's flurries faster.

Yes I have to agree on that.
I think the previous spreadsheet (by deathwing) did that, no?

Wow, you have no idea what you are talking about. You don't even need theorycrafting to realize that your MH auto attacks have far more than a 9% miss chance. It looks like you totally misunderstood the 9% hit information that is spouted on the wow forums. There is an 8.6% chance for your special attacks (yellow damage) to miss, not all MH attacks. Glancing blows cannot be decreased with hit rating either.