The prog keep writting "Could not find the hour of the day" for some sngs but I have now a result

I lost 175$ whereas I should have won 95.. playing 3.40$ sngs.. damn variance :P

Anyway, thank you very much

Could you take a hand history that produces an error and run the program on it, then look for the matching converted hand history from in the folder 'TempConvertedStarsHands' and post that, as well as the original hand history?

For example, this one :

http://www.2shared.com/file/3318946/...T88656285.html

I used your AHK script to convert it, cause starsconverter didnt work.

is that fine ?

I may have miss understood what you want me to do. Tell me if I'm right..

Hi,

Yes that is what I meant. When you run juk's program on that do you get an error? For me that converted hand history works fine.

If not could you look at the errors given by juk's program to find a tournament where there error occurs, and do the same thing with that?

How many times does it give the error regarding the time of the day? Does it say that some SNGs are missed out due to this?

it says that some SNGs are missed out due to missing final hands that were ignored

I used Juk's program with only this HH > LINK and the error with the hour of the day occurs.

I've recorded a video of what happens when I launch RunParty.bat. LINK (716ko)

Juk, ty for all your work.
I'm still interested for an update about 18 men tourneys when you'll have time/motivation (I'll pay for it since your program is really helpful).

Eventually I'll look into re-doing the ICM code as to be able to add this it needs to be able to allocate split-prizes properly and atm it's uses some really messy code to do it for 3 prizes.


As for using the GNU pokersrc library then I think I've found a good alternative. I think I can use Steve Brecher's hand evaluator library instead:

http://stevebrecher.com/Software/software.html

He's not licenced it under GNU: "Hold’Em Showdown and its source code are in the public domain (not copyrighted)." so I can see no problem with using it and not releasing the source of my whole program. It might be a little slower than pokersrc, but for this sort of thing it'll hardly have any effect.

Juk

Hi,

Neither of those links seem to work. I get occasional missed SNGs, less 1%. Do you have more than this? If not there may not be any further problem.

Thanks.

I used the program with 529 HH :

numSNGs : 502 / numdatapoints=3466

Note : 12 SNGs <47 data-points> with missing final hands were ignored

Its no big deal..

The links used to work fine, maybe it will work again later.. If they don't I can upload them on another website.

Downloadlink down again?

Yep, looks like a server crash as the other sites hosted on it have gone down too. It might come back on it's own in a bit or else you'll have to wait for it to get fixed (my lil brother looks after it and he's afk/sleeping atm...).

Juk

It's up again. Thank you!

LINK

When I have converted my ftp-sng's (these are three) and run the batch-file it keeps saying 'could not find day of the week, while the days are in the converted HH.

Any Idea?

Juk, anyway to use the luck tester for 210s ?

I had the same problem. Your windows language is not english right? If not you should try to run the program on an english windows.

I indeed have a Dutch Windows, but my time and date settings are English (US). When I played on Stars, that was the only thing i should change. Now I have to change my whole OS?

EDIT, forget it, solved

vanky just get a new ahk hand converter script (see previous page from this thread, someone put the new AHK script for pokerstars, i think it the same method pour le FTP COnverter ahk)

Is there anyway that this could be used with betfair? A hand history convertor or anything?
Also have you spoken to either the Holdem manager guys, or the poker tracker guys about getting them to add this to their software? Would be awesome imo.

Hi Juk, i have a question. I probably do something wrong because i can't confirm what your program says.

I analyzed ~4200 SNG and this is the graph:


I actually never ran that bad, i don't have a -6k downswing and then i just wanted to know how this is possible. I think that i probably did some mistake with the batch, so if you can take a look... i'm using this:

@echo off

echo Converting FTP hands to Party hands.
mkdir TempConvertedStarsHands 2> NUL
del /Q .\TempConvertedStarsHands\*.txt 2> NUL
ftpconverter.exe .\TempConvertedStarsHands .

echo Running on all 9-man FTP hands found in the TempConvertedStarsHands folder...

rem For Turbo 6s we pay 12.1212121212% of the final prize-pool as our entry fee.
BackTestAllInLuck.exe 12.12121212 50 30 20 TempConvertedStarsHands

pause


The hands i analyzed are 11+1 22+2 and 33+3 so the batch should work for everyone of them right?

Thanks for your help

Question regarding the "x SNGs with missing final hands were ignored" problem.

I compared a hh with the problem stated above to a hh which was read by the program correctly. I can't see any difference betweem the two final hands, so i don't understand the problem.

Heres is the good one:

[PHP]#Game No : 6555861302
***** Hand History for Game 6555861302 *****
NL Texas Hold'em $12 USD Buy-in Trny:49831187 Level:1 Blinds(600/1200) - Sunday, May 25, 07:18:42 ET 2008
Table Speed #1337 (Real Money)
Seat 1 is the button
Seat 1: mrjointDK ( 11,060 )
Seat 3: PetesNX ( 2,440 )
Trny:49831187 Level:1 Blinds(600/1200)
mrjointDK posts small blind [600]
PetesNX posts big blind [1]
** Dealing down cards **
Dealt to PetesNX [ 8d Jh ]
mrjointDK raises [3000]
PetesNX calls [1,240,]
** Dealing Flop ** [ Th, 7d, 7c ]
** Dealing Turn ** [ 3h ]
** Dealing River ** [ 8c ]
mrjointDK shows [ Qs, Tc ]
PetesNX shows [ 8d, Jh ]
mrjointDK wins 2359 chips
mrjointDK wins 4,880 chips
[/PHP]

And the hh with "missing final hand":

[PHP]#Game No : 6595271060
***** Hand History for Game 6595271060 *****
NL Texas Hold'em $12 USD Buy-in Trny:50126872 Level:1 Blinds(60/120) - Wednesday, May 28, 04:01:21 ET 2008
Table Speed #1337 (Real Money)
Seat 9 is the button
Seat 2: gerdim ( 1,965 )
Seat 3: PetesNX ( 775 )
Seat 7: jbeaty180 ( 2,695 )
Seat 8: jinbong88 ( 3,110 )
Seat 9: Pergamon ( 4,955 )
Trny:50126872 Level:1 Blinds(60/120)
gerdim posts small blind [60]
PetesNX posts big blind [120]
** Dealing down cards **
Dealt to PetesNX [ 7c As ]
jbeaty180 folds
jinbong88 folds
Pergamon folds
gerdim is all-In [1905]
PetesNX is all-In [655,]
** Dealing Flop ** [ 7h, 3d, 3c ]
** Dealing Turn ** [ Qh ]
** Dealing River ** [ 8h ]
gerdim shows [ Th, 9h ]
PetesNX shows [ 7c, As ]
gerdim wins 1190 chips
gerdim wins 1,550 chips[/PHP]

Any idea? What is the problem with the second one? How could it be fixed?

I have the same kind of problem as rebelzzz, I ran it at 32 games I played yesterday and it tells me I won $320 where I only won $50.

Running it over all my(165) $60's it is telling me

actual winnings: $3600
Adjusted winnings: $2400
EV luck +$1200

Where my actual winnings are $2400

I used the ftp converter and then ran the party 11-77 batchfile

Any ideas about why this is the case? thnx!.

My last 5 games

I won 2 (#xxxxx523 and #xxxxx284), lost 2(#xxxxx688 and #xxxxx918 and got one 2nd (#xxxxx807)

Note that it is not possible to have 212.7 actual winnings in a 55+5 9 player sng on Fulltilt.

net profit for first: $187.5
net profit for second: $88.5
net profit for 3rd: $39
net loss for 4-9 -$60

It got my net loss right on the games I lost, but somehow it assumes I win $212.7 when I win one and it assumes I win $104 when I get second

If you are running the Party 11-77 batchfile then probably the numbers won't add up. You need to work out what the "entry fee percentage" is for a FT $55+5 9-man (see the readme file for example).

Just got back home and very tired, so off to sleep and will reply to other posts later...

Juk

I understand it now, I figured what you meant by
entry fee %= rake/total buyin, but
entry fee%= total buyin/total prizepool

thnx!