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

[alonalbert]

Looks like you're generally getting it in ahead but you have also been running rather well except for 2 bad beats late in big tourneys (at least it seems that way but I can't tel exactly in this image)

The problem with looking at all blind levels is the big jumps you see flatten the graph for the lower levels.

You are taking the worse of it lately though but nothing to get excited about. When you were running good, it was more pronounced that your current bad run

Also, It all depends on the type of games you play too so you should really be looking at SnG separately from MTT's and 90 man games separately from large field games or filter by blind level (or both)

[corbindallas]

Looks like you're generally getting it in ahead but you have also been running rather well except for 2 bad beats late in big tourneys (at least it seems that way but I can't tel exactly in this image)

The problem with looking at all blind levels is the big jumps you see flatten the graph for the lower levels.

You are taking the worse of it lately though but nothing to get excited about. When you were running good, it was more pronounced that your current bad run

Also, It all depends on the type of games you play too so you should really be looking at SnG separately from MTT's and 90 man games separately from large field games or filter by blind level (or both)

so this app will not tell me:

#1. an app that will show me the amount of time i got it in ahead and won or lost.

#2. the amount of times i got it all in behind and won or lost.

example: "you got your money in with 65% equity on average and won 66% of the time". or "you got your money in bad 45% of the time and won 45% of the time".

[alonalbert]

Well, it kinda shows you both. Even when you get it in good and win, your running above expectation. It's not like every time you get it in good, you should be winning. Say you get in 5 times with AA vs KK and win 4 times and loose once. You're running neutral. Say you played exactly 100 hands and all of them were AA vs KK and you lost 20 of them. If you counted like you suggest, you would assume you ran really bad. You always get it in good and you had 20 bad beats vs 0 suckouts. You must be the unluckiest guy on earth. In fact though, you're running exactly neutral...

To get a feel for suckouts vs bad beats, switch to the Hands tab and sort by the Difference column. On top, you shoud have a bunch of red rows and on the bottom, you'll have green rows. The red's are beats and the greens are suckouts. Still the best measure is the graph.

From a luck point of view, it doesn't matter if you got in ahead or behind.
From skill point of view, your green line should be steadily rising...

[corbindallas]

Well, it kinda shows you both. Even when you get it in good and win, your running above expectation. It's not like every time you get it in good, you should be winning. Say you get in 5 times with AA vs KK and win 4 times and loose once. You're running neutral. Say you played exactly 100 hands and all of them were AA vs KK and you lost 20 of them. If you counted like you suggest, you would assume you ran really bad. You always get it in good and you had 20 bad beats vs 0 suckouts. You must be the unluckiest guy on earth. In fact though, you're running exactly neutral...

of course... i realize AA beats KK 80% of the time.

[corbindallas]

Well, it kinda shows you both. Even when you get it in good and win, your running above expectation.

no. this is not correct. is this how your app plots the graph? cuz if it is... it's way wrong.

when you get it in good and win you are not running above expectation. if you win 90% of the time when you are a 80% favorite... THEN you are running better than expectation.

if your app thinks winning 70% of the time when you are a 70% favorite is above expectation... that explains alot.

so it seems this app looks at each individual hand and plots a point on the graph. if you're getting it in as 80% favorite and getting called... and winning 80% of the time... holdem luck thinks you are running like a god. if you're getting it in as a 55% favorite and winning 55% of the time.... holdem luck thinks you are running just ok and the blue and green lines will be closer together.

getting all the money in... and winning the expected amount is not running good. getting all the money in as an 80% favorite and winning 80% of the time is not running better than getting all the money in as a 55% favorite and winning 55% of the time.

i'm curious... what do you mean by kinda? either it shows you both or it doesn't... if it "kinda" shows you both...then it's only answering part of the question... or adding information to skew the answer.

[alonalbert]

No, you're interpreting what I say wrong.

If you play a single hand where you equity for that hand was 80% and won the hand, you are indeed running above expectation for that hand. You were expected to win 80% of the pot and you won 100% of it.

If you win 80% hands 80% of the time, the app will show you running neutral which exactly what you are.

For every single hand, the app calculates your expected value which the equity * pot size and compares it the amout actually won/lost. if it's higher,you ran good on that hand, if it's lower, you ran bad. It's that simple. For every hand, the red line will either go up if you ran good or down if you ran bad. Even for hands where you were the favorite and won, the graph will go up just a little bit.

Example, if you win 4 80% hands in a row and then loose one, the graph will go up a little bit 4 times and drop like a rock on the 5th hand and will end up at zero. Trust, I simulated such hand sequences including Rake for cash hands to verify it works properly.

[alonalbert]

By Kinda Both I mean:
It doesn't show you how you run when get it in ahead vs how you run when you get in behind which is what you're asking.

In my opinion, it doesn't matter if you got it in ahead or behind, it show how you run period.

The direction of the green line shows you if you are making good decisions.

[corbindallas]

i would bet that holdem luck is skewed toward giving more weight when you run as expected... for example winning 80% equity pots 80% of the time. cuz there is no way anyone runs as supposedly good as i've run over 270k hands. my OPR ranking is in the 99th percentile. i'd like to see other winning players graphs... i bet theirs looks like mine... with the blue line ahead of the green line. and i bet losing players have the blue line below the green line. because good players get their opponents to get their money in behind.... consistently.

however... no one runs good consistently over 270k hands.

in fact... i probably have 50k hands before i bought poker tracker... the up turn in the beginning of my graph was right at the point i got some excellent coaching after i'd been playing about 6 months.

i'd like to see graphs from other winning players... i bet they "run rather well" over their sample size also.

[WhiteRider]

i would bet that holdem luck is skewed toward giving more weight when you run as expected... for example winning 80% equity pots 80% of the time.

You can't win one pot 80% of the time - you either win it or lose it.
If the pot is $100 and you have 80% equity then your expected win graph will go up by $80, but in reality you will either win $100 or $0 so the actual win graph will either go up by $100 (above expectation) or $0 (below expectation). (in simple terms)

[marvin]

it shows a lot of hands twice (also if checking on PT3, the hand appear only 1 time) i have tried to vacuum-cluster the PT3 DB (may be a duplicated hand on import process of PT3) but no works. is a bug of the v.18? it is fixed on v.19?