Latest cliffsnotes on Absolute soulreading.

[Josem]

OK.

I've just read the wikipedia entry on standard deviation, and am more convinced than ever before that flight2q has no idea what a standard deviation is.

The idea of calculating a standard deviation on a single result from 190 hands is absurd.

A standard deviation is a measure of a spread of results. Inherently, a single result obviously can't be spread.


So, I took the bb/100 data in excel, and firstly found the mean of the 5251 results. The mean was -4.0493. That is, of the 5,251 players who played over 500 hands, they lost on average 4bb/100.

I then used MS Excel's STDEVA function to calculate the standard devation of that same sample ('cause I certainly wasn't doing 5,000+ calcs by hand!) and it reported that there was a standard deviation of 31.54.

Now, the cheater's BB/100 was 475.74. That is a whopping 15.2 standard deviations from the mean.

To put that into perspective, Wikipedia lists confidence intervals for standard distributions that assume normally distributed data:

σ 68.26894921371%
2σ 95.44997361036%
3σ 99.73002039367%
4σ 99.99366575163%
5σ 99.99994266969%
6σ 99.99999980268%
7σ 99.99999999974%

They don't even bother going past 7 standard deviations because of how rare such occurences are.

[flight2q]

(Added: About your post just above this one, you have no idea what you're doing. You are just throwing together random numbers and think they mean something.)

[ QUOTE ]
[ QUOTE ]
Guess Holdem Manager doesn't report S.D. yet. That blue dot loses close to 100 buyins in 2783 hands. Not easy to do, especially with 22% VPIP.

[/ QUOTE ]

That "blue dot" lost $142.10. They seemed to play mostly 10NL at stars.

[/ QUOTE ]And yet it must be close to 100 buyins, if the value you plotted for the bb/100 is correct.

[ QUOTE ]
[ QUOTE ]
Red dot has S.D. of 520bb/100 calculated from a different 190-hand sample (deep stack) for one of the accounts.

[/ QUOTE ]
I am fairly confidence that is not what "standard deviation" means.

[/ QUOTE ]You don't seem to know what standard deviation is.

[ QUOTE ]
[ QUOTE ]
Red dot typical preflop raise is to 3bb.

[/ QUOTE ]
I don't know where this data claim is from.

[/ QUOTE ]It came from examining hands played.

[ QUOTE ]
[ QUOTE ]
Average first bet put into pot, when not reraising, is prolly around 2.5bb.

[/ QUOTE ]
I don't know where this data claim is from.

[/ QUOTE ]It comes from the above, plus the cited vpip/pfr stats.

[ QUOTE ]
[ QUOTE ]
Playing say 4-handed on average,

[/ QUOTE ]
I don't know where this data claim is from.

[/ QUOTE ]It comes from examining hands played.

[ QUOTE ]
[ QUOTE ]
the equity loss due to too loose preflop shouldn't be more than 1bb per hand (50PTBB/100).

[/ QUOTE ]
I don't know where this data claim is from.

[/ QUOTE ]This is a very rough estimate of the equity of a person putting in 2.5bb and being called by the best of 3 other hands. Looks like an overestimate.

[ QUOTE ]
[ QUOTE ]
Actually, clearly less than this, because steals successfully quite a bit and play is positional from what I saw.

[/ QUOTE ]
I don't know where this data claim is from.

[/ QUOTE ]This comes from looking at hands played.

[ QUOTE ]
[ QUOTE ]
Nice scatter plot. But usually number of deviations from expectation is what is plotted when different samples have different S.D.'s.

[/ QUOTE ]
wtf?

the sample is the bb/100 results and the VPIP results.

there are two samples here.

[/ QUOTE ]You don't seem to know what sample means.

[jason1990]

Poker Tracker does not calculate standard deviation correctly. It uses only the results of each session, and not the results of each hand. This has two effects. The major effect is that the SD it calculates can be grossly inaccurate if the number of sessions is not large. The secondary effect is that the SD it calculates can be biased if session length is at all correlated to the session winnings.

If you want to correctly calculate the sample standard deviation of the alleged cheater, then you must compile an Excel spreadsheet that contains the results (in BBs) of each hand played by the alleged cheater. If you have 500 hands played by the accused, then you will have a column of 500 numbers. The sum of these numbers will be the total BB won by the accused. The average of these numbers will be his sample winrate in BB/hand. Multiply that by 100 to get his winrate in BB/100.

Now use Excel's STDEV function on this column of data. The result will be his sample standard deviation in BB/hand. Multiply it by 10 to get his SD in BB/100.

[flight2q]

jason1990, the S.D. was calculated as you describe using a 190-hand session of one of the suspect accounts provided to me (this hand history fails to import into PT). The S.D. was around 260 PTBB/100. Since there are some rather large pots in deep stack games, the 500 hands that appear to be successfully imported into PT might have a substantially higher or lower S.D.

[jason1990]

Well, that kind of SD is obviously not sustainable, for cheaters and honest players alike. If your calculation is correct, it simply suggests that the sample size is too small for an accurate measure of SD. It can take a "normal" player about 5000 hands or more before their SD stabilizes. A crazy player can take even longer. Obviously, any attempt to compute his SD and to draw conclusions from that computation is futile without many, many more hands.

[adanthar]

***breaking news***

incoming "beyond a shadow of a doubt" proof in under 24 hours. it will be in a new thread.

sneak preview: absolute [censored] up and sent out the master HH of *every hand played* in the Potripper tourney

[TheWacoKidd]

[ QUOTE ]
***breaking news***

incoming "beyond a shadow of a doubt" proof in under 24 hours. it will be in a new thread.

sneak preview: absolute [censored] up and sent out the master HH of *every hand played* in the Potripper tourney

[/ QUOTE ]

*ahem* you're welcome

problem is, the file is all jumbled and we are having trouble getting a full table into the replayer... at minimum we will copy and paste numerous key hands for everyone to see

and WTF adanthar use your mod powers and get rid of this [censored] newbie [censored] next to my name lekfrsknlglk

[augie_]

[ QUOTE ]
***breaking news***

incoming "beyond a shadow of a doubt" proof in under 24 hours. it will be in a new thread.

sneak preview: absolute [censored] up and sent out the master HH of *every hand played* in the Potripper tourney

[/ QUOTE ]

[bubaloo]

[ QUOTE ]
***breaking news***

incoming "beyond a shadow of a doubt" proof in under 24 hours. it will be in a new thread.

sneak preview: absolute [censored] up and sent out the master HH of *every hand played* in the Potripper tourney

[/ QUOTE ]

on a scale of exiting implications 0 being ATF and 10 being a 1.4m hand fgators ev graph, i would say this is an 8.3.

after some consideration, im gonna move it up to a 9, this is pretty big.

[moving shapes]

somebody is getting fired.