The Official FGators Put up or Shut up Thread - Page 54

prodonkey · #**531** 10-11-2007, 07:48 PM

[ QUOTE ]
I don't see how this can live up to the hype. The only way it can is if he was leveling us all along and he's up like 100k.

[/ QUOTE ]

not a chance he's up 100k... not a chance he's in the green

_dave_ · #**532** 10-11-2007, 07:50 PM

[ QUOTE ]

we need to find a way to make PT handle the data.

[/ QUOTE ]

You will obv. have to be using postgres database.

Split the files into ~100 hand chunks (maybe 50 hands/file would be better - but double the clicks, import.

Or import in big chunks, and use the program I linked above to keep an eye on the import process (if it works, that is)

PT will almost never crash on import - it just looks like it has. Larger HH files cause disproportionate increasing import time, thus why splitting is sensible.

fluorescenthippo · #**533** 10-11-2007, 07:50 PM

[ QUOTE ]
I don't see how this can live up to the hype. The only way it can is if he was leveling us all along and he's up like 100k.

[/ QUOTE ]

yea his losses will just be like withing 1% of his expected losses. the only cool thing will be if it will be possible to show that his style is causing his loss.

ginko · #**534** 10-11-2007, 07:51 PM

Wait I thought you said you had a 50k roll in another thread..

donkeykong2 · #**535** 10-11-2007, 07:51 PM

so here is what FGators can expect realistically:
I dont know his standard deviation/100 so i assume it is around 60bb on average.
he has 4 mio hands, thats 40000*100.
the standard deviation rises with the square root of numbers.
so thats 40000^0.5*60bb=200*60bb=12000bb
12000bb is one standard deviation. that is around 240 buyins.
if he ran normal (95% probability) he should be within +-2*240buyins, so 500 buyins above or below expectation. it is just an approximation as he jumped levels so the math behind it is not 100% correct. say average stakes are 200nl:
500*200=100000
so he should be around +-100000$ from his expectation.
im not sure if the sd/100=60bb is an ok assumption though. pokerev will also only cover a part of the variance. so 100000 should be an upper limit under these assumptions.

whoisthedrizzle · #**536** 10-11-2007, 07:52 PM

Fgators, do you coach?

prodonkey · #**537** 10-11-2007, 07:52 PM

[ QUOTE ]
Wait I thought you said you had a 50k roll in another thread..

[/ QUOTE ]

Butcho22 · #**538** 10-11-2007, 07:52 PM

In on the middle floor

hra146 · #**539** 10-11-2007, 07:53 PM

I dont even know myself how much hes up or down. I imported like 50k just to try out HM and since then its importing the entire (what I believe is no more than 2m hands) in little pieces. [Little pieces being about 10k hands every 1m10sec]

prodonkey · #**540** 10-11-2007, 07:54 PM

[ QUOTE ]
so here is what FGators can expect realistically:
I dont know his standard deviation/100 so i assume it is around 60bb on average.
he has 4 mio hands, thats 40000*100.
the standard deviation rises with the square root of numbers.
so thats 40000^0.5*60bb=200*60bb=12000bb
12000bb is one standard deviation. that is around 240 buyins.
if he ran normal (95% probability) he should be within +-2*240buyins, so 500 buyins above or below expectation. it is just an approximation as he jumped levels so the math behind it is not 100% correct. say average stakes are 200nl:
500*200=100000
so he should be around +-100000$ from his expectation.
im not sure if the sd/100=60bb is an ok assumption though. pokerev will also only cover a part of the variance. so 100000 should be an upper limit under these assumptions.

[/ QUOTE ]

Very wrong.. this would mean that a win rate never converges.. and that playing more hands means you're getting farther and farther away from any kind of true measure of expectation.