monk ratios, primary and offhand choices

[Jimjam]

Attack vs mitigation is a roll off, ie two dice, which is why it creates a peak near the centre. Like how two 6 sided dice have a sum of 7 as the most common result.

The peaks which can be seen at the min and max are due to the roll off producing results that are out of bounds so they get lumped in as min or max hits as appropriate. That is my current understanding of it.

[bcbrown]

Originally Posted by Jimjam [You must be logged in to view images. Log in or Register.]

Attack vs mitigation is a roll off, ie two dice, which is why it creates a peak near the centre. Like how two 6 sided dice have a sum of 7 as the most common result.

The peaks which can be seen at the min and max are due to the roll off producing results that are out of bounds so they get lumped in as min or max hits as appropriate. That is my current understanding of it.

This is plausible. But the fact that the damage values close-to-half are not higher than average, while the exact half has a strangely higher count, leaves me disinclined to accept that explanation.

Here's the raw data. The first column is the number of times a hit of that value occurred, the second column is the damage.

First, mainhand punch:

 

10 10
1 11
2 12
2 13
5 14
6 15
5 16
7 17
8 18
3 19
6 20
2 21
8 22
5 23
3 24
7 25
7 26
16 27
2 28
3 29
1 30
6 31
1 32
4 35
2 36
1 37
1 38
1 40
2 41
1 42
1 43
1 44

Offhand SoS:

 

5 2
1 4
3 5
3 9
1 10
2 13
3 14
8 15
1 16
2 17
1 18
2 19
1 20
1 22
2 23
2 24
1 26
2 28
3 29
1 30
1 31
1 32
9 34
1 36
3 37
1 39
1 41
1 42
1 46
1 47
1 48
2 49
1 53
1 55
3 57
2 59
1 66

Mainhand IFS:

 

6 22
1 29
1 33
1 45
1 48
2 52
1 53
1 54
1 56
1 60
1 62
1 64
1 65
3 67
1 69
2 71
1 75
1 77
2 79
1 83
1 85
1 86
3 94
2 97
1 99
1 100
1 101
1 102
1 114
1 119
1 123
1 129
1 132
1 133
1 134
1 140
1 151
1 154
1 162
1 169

Offhand punch:

 

4 1
1 3
2 5
2 7
1 10
1 11
2 12
1 14
1 15
3 16
1 17
10 18
1 19
1 20
2 21
1 22
1 24
1 26
1 30
1 31

You can see that on all of them the rolls are biased towards lower hits. But separately there's also big spikes for punches at the halfway value, a moderate spike for SoS, and no spike for IFS.

[Troxx]

Or you can acknowledge that the parse is too short for actual patterns to reliably emerge bcbrown or so short that false patterns that don't exist could coincidentally be observed. We see such false results often in medical clinical trials with too few patients in them

2-3 mins is really short

[DeathsSilkyMist]

Originally Posted by Troxx [You must be logged in to view images. Log in or Register.]

Or you can acknowledge that the parse is too short for actual patterns to reliably emerge bcbrown or so short that false patterns that don't exist could coincidentally be observed. We see such false results often in medical clinical trials with too few patients in them

2-3 mins is really short

Troxx accepts Ripqozko's 2-3 minute parse without question. He keeps forgetting that.

Troxx could provide his own parses at any time!

[Troxx]

There are too few HITS to get an even spread on distribution to extrapolate the interaction between mob ac and player attack.

Jesus fucking Christ go stick your head in a toilet you fucking baboon. Let the big boys with big boy brains have an intelligent conversation for once FFS.

[DeathsSilkyMist]

Originally Posted by Troxx [You must be logged in to view images. Log in or Register.]

There are too few HITS to get an even spread on distribution to extrapolate the interaction between mob ac and player attack.

Jesus fucking Christ go stick your head in a toilet you fucking baboon. Let the big boys with big boy brains have an intelligent conversation for once FFS.

You claim to be a big boy with a big brain, yet you never show us either of those things lol. We need to pay 150 dollars for the priveledge of your logs for somw reason. People can use all of my videos and logs for free.

Show us your big boy logs, show us how it's done!

[bcbrown]

Originally Posted by Troxx [You must be logged in to view images. Log in or Register.]

Or you can acknowledge that the parse is too short for actual patterns to reliably emerge bcbrown or so short that false patterns that don't exist could coincidentally be observed. We see such false results often in medical clinical trials with too few patients in them

2-3 mins is really short

Oh absolutely agreed. I use the data to which I have access, not the data I wish I had. All this parse DSM provided can provide is hypotheses to either be confirmed or rejected by a parse of a proper length. But I've seen this spike at the middle in other logs as well. At this point my belief has changed from it being an artifact of too-short logs to being a real effect, probably to do with compressed DI on low-damage weapons.

[DeathsSilkyMist]

Originally Posted by bcbrown [You must be logged in to view images. Log in or Register.]

Oh absolutely agreed. I use the data to which I have access, not the data I wish I had. All this parse DSM provided can provide is hypotheses to either be confirmed or rejected by a parse of a proper length. But I've seen this spike at the middle in other logs as well. At this point my belief has changed from it being an artifact of too-short logs to being a real effect, probably to do with compressed DI on low-damage weapons.

Feel free to provide better data at any time lol. People keep complaining, and never solving their own issues.

[Troxx]

Originally Posted by bcbrown [You must be logged in to view images. Log in or Register.]

Oh absolutely agreed. I use the data to which I have access, not the data I wish I had. All this parse DSM provided can provide is hypotheses to either be confirmed or rejected by a parse of a proper length. But I've seen this spike at the middle in other logs as well. At this point my belief has changed from it being an artifact of too-short logs to being a real effect, probably to do with compressed DI on low-damage weapons.

It is not a pattern typically observed either offensively or defensively. Defensive parses typically have a more smoothed out bell curve of sorts unless you have ac/gear that completely eclipses and overpowers the content. Offensive parses have always been more scattershot for sure but this type of spike is likely erroneous and due to the short parse length. Short parses , on average, will give you an average … but they are prone to some really wonky results at times.

Originally Posted by ResidentFuckwad

Feel free to provide better data at any time lol. People keep complaining, and never solving their own issues.

Shhhhhh … the grown ups are talking now.

[DeathsSilkyMist]

Bcbrown, ask Troxx for logs. He claims he will give logs out to anyone but me for some reason. You can get the data you need!