Does accuracy not scale linearly? Question for those with lots of turn 5 XP.

[WutangRZA]

So while we've all discovered that you need more accuracy in turns 1 and 2 than you do in primals, and you need more accuracy in turn 4 than in turns 1 and 2, and finally that you need more accuracy in turn 5 than anywhere else in the game, has anyone else noticed that accuracy does not seem to scale linearly and that accuracy capping may in fact represent a DPS loss?

We finally killed Twintania last night, and over the weeks, though my ilvl has gone up, my accuracy has actually gone down. When I had > 490 accuracy buffed, I of course never missed on turn 5. When I fell down below 470 buffed, I started recording misses, but maybe only .1-.2% of the time.

[WutangRZA]

Fast forward to our kill last night, and I'm down to 447 accuracy unbuffed and with HQ deviled eggs across 2 long attempts and a kill, I only missed around .1-.2% of the time.

Has anyone else noticed a similar scaling of accuracy on turn 5? Is it possible that accuracy scales linearly up to a certain point and then slopes off rapidly over say 450 or so? Weeks of turn 5 attempts lead me to believe this is the case, but I don't really want to start preaching it as gospel, especially if others aren't seeing the same thing.

[Adam]

So you're hypothesis is: after a certain point it's better to use other stats besides focusing on making sure your accuracy is capped?

I'm I correct in understanding what you wrote?

[Garnatian]

You have the parses to back up those accuracy %s?

[WutangRZA]

So you're hypothesis is: after a certain point it's better to use other stats besides focusing on making sure your accuracy is capped?

I'm I correct in understanding what you wrote?

That's right, unless of course you're a BLM or something where a single miss is going to hose up everything for you.

But if it's true that accuracy doesn't scale with constant returns, and instead gives say linear returns up till some point and then slopes off rapidly, that means people should probably, in general, focus less on accuracy.

Starting next week, I'll start doing a better job saving data on turn 5, but over multiple attempts at around 450 accuracy unbuffed, I never seemed to miss more than .15% of the time or so on Twintania

Maybe starting next week, turn 5 groups can start posting some parse info on attempts and we can tabulate it all.