Question about avoidability

The 'Let's get critical'topic got me wondering.

If there's a stat that can enhance a characters critical rate to a 100%, would this theoretically also be true for avoidability?
I've noted this in the other topic, but nobody responded so I decided to make a new topic about it.

Imagine, a 100% avoid rate. Would be pretty cool.
However this is just an idea, I've got no clue how exactly avoidability works, or what influences it.

Could someone help me out?
What are the factors, what could be the calculation etc.

I know that LUK, an the targets ACC has to do with it... But other than that, i do not...

Kinda wondering this myself, I dont think monsters can every have a reduced chance to be crit. But I think their accuracy based on level will never allow perma avoid.

Avoidance is a tricky one.
The evasion stat is calculated from luk and dex, but the resulting number is not a %.
I have a suspicion that the enemy level can impact avoidability.

Yes I think so as well,

I reckon it would be something along the line that the amount of LUK and DEX + any avoind boost by equips and/or skills would result in a given value.
And that this value should be higher than a given monsters ACC.

This would theoretically mean that your avoid should be higher than the highest ACC value of any monster in the game.
The tricky part is that his is not completely true, even the lowest level monsters occasionally land a hit sometimes...

Do you know how to get the accuracy values of enemies?

No, not really. I knew that in MS there were vast databases for this, I doubt that those values would be the same though.

What characters do you have? And what are their evasion stats?

Would be good to have a high level DB to test with...

I've got a DS, with 38 LUK avoid of 169.
If I'd have to guess, the level 125-127 monsters of ToT hit for about 30%-40% of the time.

But This is just guessing, I would like to know an exact formula or how to work this out.
But I hardly have a clue what all the variables are.

Well, we dont know the formula so we have to figure it out through testing.

Can you get a bit more exact on that number?

E.g. tested n=120 times against lvl 125 mobs resulting in a hit=x and miss=120-x

Yeah I'll try to work that out .

Cool. I assume you are 150?

I will do some testing with my AB later. Pretty sure I have a different evasion stat.

I'm only 139, and probably not going to get to 150. At least not if the last 10 lvls are manual play in those bullshit levels like MK and LHC. Then I'm quitting.

okay so I've done some field testing:
Hits 49 : 64 Misses Lyka lvl 127 @169 avoid (113 total)
43,3% Hit  56,6% Miss.

Hits 25 : 104 Misses Tino lvl 1  @169 avoid (129 total)
19,3% Hit 80,6% Miss.

These are results so far

Evasion 124 ; Level 150 ; AB

Enemy Lvl 1 ; Hit:Miss 17:103 ; 85.8% miss

Enemy Lvl 48 ; Hit:Miss 42:78 ; 65% miss


Hard to make much of it yet...

Yeah, it doesn;t really make sense..

Your evasion is worse than mine, yet your percentages seem equal or even a little better than mine...

Maybe it's also different for each class?
Or perhaps it has to do with some kind of character level vs. monster level factor as well, since the only other difference would be theat you're lvl 150 and I'm only 139...


I played around a bit with the MS calculator on Hidden Street as well... But I'm not sure if this is correct

Also I found this explaination on BasilMarket:

'There's four types of Avoid:
~Avoid in your stat window. This is used to calculate Evasion Rate.
~% Avoidability. This is just a multiplier for the Avoid in your stat window. An example is Katara Expert with 15% Avoidability. You have 5000 Avoid in your stat window, and multiplying that by 1.15x = 5750 Avoid.
~Dodge Rate. This is typically found in skills with a flat chance to activate, resulting in you taking no damage.
~Evasion Rate. There's two parts of this. One is found in skills, but is a multiplier on top of the Evasion Rate you calculate. The other is basically your chance to trigger the purple 'MISS'/'GUARD'. The formula to calculate that is listed in the link above, and note that Avoid Rate is the same thing as Evasion Rate.

So for simplicity's sake, you have 5000 Avoid in your stat window. Square rooting that becomes 70 rounded down. An enemy you face has 500 Accuracy. Square rooting that becomes 22 rounded down. Your Avoid - Enemy's Accuracy = Your Evasion Rate, so 70 - 22 = 48% Evasion Rate.

Considering you have a DB, we'll include their 3 Evasion Rate skills (Side Step, Shadow Meld, Mirrored Target). As noted previously, Evasion Rate skills are a multiplier, so 3 of them add up to 70%, which is 1.7x. Using the previous 48% Evasion Rate, and multiplying that by 1.7x = 81.6% Evasion Rate. That means you have a 81.6% to trigger the purple 'MISS'/'GUARD'. Also, Evasion Rate caps off at 90% post-Unleashed, so anything over 90% is ignored.

Now let's say you have 50% Dodge Rate from a skill, and your Evasion Rate is still at 81.6%. 100 - 81.6 = 18.4% to get hit by an enemy. Of that 18.4%, you have a 50% chance of Dodge Rate activating, which is 9.2%.

In summary, you have 81.6% Evasion Rate, and an additional 9.2% Dodge Rate to evade an enemy.'

Is that for the pc maple?
I think the formulae are different for pms.

As to our samples, there could easily be a +/- 5% deviation from expected rates.

It would be good to test at the extremes, i.e. very low (5%) avoidance, or very high (95). If that is at all possible.

I don't think the formula changes with class, but its hard to say at this stage.

Yes that is correct, I do think so as well that hose formulas don't apply directly to PMS, however I do think they give a representation of all variables included.

The only thing is... how do we manage to get the ACC from enemies?
We could just state an imagitive maximum, but that doesn't really seem representable.

And I agree we need bigger sample groups.

few more samples...

AB
Evasion 5
Level 9
Enemy 1
n = 120 hit = 110 miss = 10 (avoid rate 8.3%)

AB
Evasion 43
Level 42
Enemy 53
n = 80 hit = 80 miss = 0 (avoid rate 0%)

So, it's starting to emerge that the key variables are Evasion stat (obviously) and Enemy Level.
I was hoping the relationship was linear, but unless the formula differs between classes, then it feels unlikely.

Hmm okay, interesting. But it's quite obvious that the relationship would be exponential. At least judging from assumption, because it already was exponential in the original pc version of MS.

So would we be able to set up a logarith from the date we have, or are we still missing some variables? (I'm extremely bad at this kind of math I must add.)

And how accurate would that be, since we havent factored in things like actual monster ACC, and character level. Hoewever I'm also doubting about the significance of the latter.

I will try to get some more samples with my DS later, if I manage to get to it.
Anyways thank you for the input!

The original pc version uses square roots, not exponential.

more datapoints would be nice to test whether the above model holds.
the other variables we have don't appear to improve the results, within our data range anyway.

i'm not even sure the enemy accuracy variable even exists in this game.
using enemy level as a proxy seems to work very well, anyway.

why do you need the log?

Square roots are derived from squares thus involve an exponential curve? or am I wrong? The original MS certainly does not have a linear equation.

What I actually mean by 'set up a logarithm' is:
'Can we create a formula that can be used to calculate the number of avoid needed to gain a permanent avoidability?'

Sorry for the confusion.

Square roots are x^0.5
Exponentials are exp(x)

The regression I am seeing shows a max evasion of 85%.

you're right haha, I never was a real mathematical genius.

Yeah, I've noticed that as well... doesn't look too good for permaavoid

I wouldnt rule it out yet.

One thing you could do is use your 169 avoid DS, drink some potions that increase avoidance, and see if it improves your dodge rate to above 90% against the lvl 1 mobs.

I was very disappointed when your DS only had 80% against them...xD

true, I could do that yeah! I haven't been active for a few days, and I don't think I can be for another week or so, so it might take some time.

I wasn't so much dissapointed as I was suprised actually, that's why I pressumed that character level would play an important role.

My DS was 139 with 169 avoid, whilst your AB was 150 with 120~ avoid