Application: HoldemLuck/TourneyLuck v0.20 (Updated:2/28/10)

[kraada]

It can take quite a while the first time, it should take noticeably less time after that.

[Clockwork Banana]

I just downloaded it, unzipped it and loaded it up...

But it has been "finding new pots" and loading for an hour now. I do have a pretty large database. How long does this generally take the first time?

Thanks.

It's very long, it take about a couple of hours, maybe 3 depending on how many hands you have. I have about 350K.

If your database has that many hands or more you might end up with the same problem as I have (see my post on page 39 for details).

But since yours has been running for 1 hour already that might not happen to you. I get the error message about 10mn into it.

How many hands do you have?

[uconnrounder]

Hi.

Sorry for posting something that I probably could find on my own if I spent enough time searching, but how does the holdemluck/tourneyluck do the analysis for sng play? Are the EV calculations for sngs ICM-based or is it simply based on expected chips?

[kraada]

I believe ICM is used, but I'll leave it for alon to explain the details as I don't know those

[gp18089]

I take the message:

Fatal error. See log for details: A timeout has occured. If you were establishing a connection increase Timeout value in ConnectionString. If you were executing a command, increate the CommandTimeout value in ConnectionString or in Your NpgsqlCommand object.

This player has 320,000 hands. To another player with 35,000 hands works fine. Any ideas?

[Clockwork Banana]

I take the message:

Fatal error. See log for details: A timeout has occured. If you were establishing a connection increase Timeout value in ConnectionString. If you were executing a command, increate the CommandTimeout value in ConnectionString or in Your NpgsqlCommand object.

This player has 320,000 hands. To another player with 35,000 hands works fine. Any ideas?

Yeah I have the same problem with 350.000 hands (see my post on page 39). According to Alon Albert when the database become too big, it does that. He had a fix for this problem but was written for Holdem Luck rather than SNG luck and he didn't have the time to write one a the time.

Since he hasn't checked this thread for a while, my guess is that he's on holiday.

We'll see what he says next time he checks the thread otherwise we'll have to partition the database and that sounds like a headache.

[uconnrounder]

Yeah I got that error as well with a 700k hand sng database...beyond that I feel like I'm doing something wrong. I downloaded the zip file and basically double clicked the SnG luck file and it searched for pots...how does it know where my database is? did I skip a step?

[kraada]

When you first open it up it should ask you, but you can check a "auto-fill data from PokerTracker 3" box. I know you can get that box again if you start the program up without PostgreSQL running.

[uconnrounder]

Yeah,

I wound up PMing Alon about how the program calculates tourney luck. It just uses chip EV, and while Alon already knew this, I don't think everybody realizes how little cEV means in a SnG. You can run horribly under cEV over very long stretches of sngs and still be running well above $ equity in a sng... In the end, it's the $$ that matters, not the chips. Additionally, it doesn't take into account how your equity changes in hands that you fold and other players get eliminated/stay alive. Is PT3 working on this? It's the only big (and I mean huge in my opinion) advantage that HEM has over PT3. I honestly can't wait to delete HEM from my computer because I think it sucks in every regard except for the equity calculations, but as I play strictly super turbos, I need the luck analysis to maintain my sanity. Just my 2 cents, and I'm sure there are others out there who agree. Thanks for the great product; an improvement in the $EV tourney luck is really the only thing I can think of to make it better.

-Rob

[Clockwork Banana]

I disagree I think cEV is most useful in SNGs (even more in DoNs) because there's a very strong correlation between the amount of chips you win and the payout you get. It's in MTTs where it's unclear.

If you triple up in the first hand of an SNG you're probably gonna win it. Tripling up in the first hand of an MTT is almost meaningless in terms of how likely you are to make the final table.