SMN BiS

[Seobit]

Torin posted a while ago showing that DET statistically parsed better than a CRIT build... by a very slight margin in the decimals. You could go either way and would probably not see the difference. Personally I think a more balanced approach is better.

That is because gear in general contains less det on it than other stats, it doesn't change the fact that 1 for 1 det is considerably more valuable than crit is. The i90 crit and det sets had 71 crit vs 45 det, and both came out almost equal, which in itself proves how much stronger of a stat det is compared to crit. I see people in this threat trading away 62 det for 67 crit in their sets which is absolutely horrendous.

My personal testing was swapping out crit heaving gear for Det heavy or just crit-less gear in T5

With allagan book and every piece of crit i90 I had I did ~340-350dps pretty often with a crit rate of 25% at best with lows of around 320 and 20% crit rates

With det and critless items (tidal wave book, moogle ribbon and some other changes but all ultimately higher int and magic damage too) I did an average of 320-330 with a high of 344 once. averaged 15-17% crit rates with this equipment. It entirely depends on how often you crit but heavy crits can outdo the sustained damage of det or underperform. Either way you actually need to test in game, not just do the math and run simulations.

On the contrary, anecdotal evidence has very little worth. The fact that you can get as much as 30 dps difference in your results from using the same exact gear shows how unreliable it is as a form of measurement. In game testing is prone to human error which invalidates the results categorically.