new evasion stat - feedback

Indi

I did some testing, and have some feedback and a question. Testing done on the exact same mob (including it being the same mob id), at 0, 50 and 250 evasion invested (150, 200 and 300 effective evasion). Test sizes of 1000 hits, except for the tests with no stats invested where I stopped after a few hundred (enough to accurately say there is no reduction, but not to speculate on the actual proc chance.) Even with those reps, these figures probably differ from the actual figures, but should be accurate enough for the comments to have some weight.

Results:
With 150 effective, evasion has a chance to process, but when it does there is 0 reduction
With 200 effective, evasion has ~12.5% chance to process, with 10.5% reduction
With 300 effective, evasion has ~18.8% chance to process, with 30.5% reduction

Question for @Tecton
Is reduction intended to be calculated based on evasion stats invested, rather than effective evasion?


Comments:
In my opinion, extremely low return on stats invested. Still the weakest stat and not worth investing in, for 2 main reasons:

1. Even averaged, the reduction is insigificant - with 50 invested and calculated over a long time, 1.3% less damage taken. With max evasion, less than 6% less damage taken over time. Other stats return much more.
2. RNG - if I can't depend on a reduction per mob fight or 20 second section of pvp, then the reduction needs to be spectacular when it happens. 

Possible solutions:

• Much higher chance to process, with significant DRs on that chance 
• Much higher reduction
• Some combination of both above
• Flat reduction in all damage, without any chance to proc (boring)

Cubey

IMO rng-based damage reduction doesn't belong in a game like Starmourn. It works for MMOs, your WoWs and FF 14s because pve lasts longer, everything takes more hits to die so the reduction averages out over time of a single group pull or boss battle.

Here, everything dies in 6 hits or less and fighting more than one mob at a time is suicide. Evasion just isn't a good option tactically: even if its effective HP increase was more in line with other stats you can still get fucked over by a string of bad RNG rolls.

Slander

I've been wondering about the Evasion change myself. I took about a month break and it came through right after I came back, I was told it was an 'all or nothing' stat and if I didn't want to max I shouldn't bother. I'm not a hardcore, optimize everything gamer but I don't want to make entirely stupid decisions, of course. Sooo... I guess I'm not adding anything terribly useful to the conversation, more asking confirmation that it is, in fact, as fucking useless and awful as it seems?

Indi

@saturnine
Well, even if you go all in, you have a less than 20% chance to get roughly 30% less damage on a single hit. Even if it literally procced once every 5 hits, that is a very low return on 250 stats. Add rng where (during the tests) it isn't uncommon to go for 20 hits without one evade, and the unreliability of it makes it even more of a waste of time.

So yes, it is rubbish. Even when you go all in.