Tight play wins the money? Pokertracker says not! - Page 3

chesspain · #21 10-23-2004, 04:28 PM

[ QUOTE ]
You honestly can't see this is a sample size issue? You've played 8000 hands. You only include people with 30+ hands. How many players is that leaving in your data? My god, you refuse to draw any conclusions about your own winrate because you only have 8000 hands, then you proceed to break down your database into god knows how many groups and examine their winrates? How many players in each category? How many total hands in each category? A few hundred max? And cutting out those with less than 30 hands serves no purpose either. So anyone with 31 hands is in huh? If he is really a 15% player he will play 4.65 hands out of every 31 on average. If he happens to play 8 hands in that 31 hand sample, you've got him as a 25% VP$IP! Wrong category. How many players you think you have in the right category? How many players do you have more than even 100 hands on? Come on, the data you have is completly useless for what youre trying to do. You cant even have any confidence in your categorization of players, nevermind their winrates.

[/ QUOTE ]

Well said!

JDErickson · #22 10-23-2004, 04:36 PM

This all comes down to sample size like everyone else said.

I decided to look into my figures because I had nothing better to do [img]/images/graemlins/smile.gif[/img]

I filtered all players with +500 hands (even this is too small to matter)

What stood out:

Maniacs, LPA's and SLAP's I had 0 players meet the requirements. No surprise here. Play this way you don't stay around long.

TAA's are doing quite well (surprise surprise) + 2.96 bb/100

LAP's do the worst at -2.95 bb/100

Statistic anomaly - TPP's +3.69 bb/100

Jdanz · #23 10-23-2004, 04:43 PM

Hate to disagree with the pros, but there is one specific reason why he got those numbers.

You have very few hands per person, therefore the people who are playing la-a are probably people who just have a good run of cards.

Why are they winning, you get a couple of high pocket pairs in a 50 hand session you get very loose very aggressive seeming and you win a lot of BBs.

-JDanz

donkeyradish · #24 10-23-2004, 05:12 PM

Makes sense! Most of those LA-A players are probably TA-A players. But because I don't often see the same people at a 1$/$2 table, and because people come and quickly go, I never find that out.

And to those who said "your sample is too small" this would still be true even if I had a million hands in my hand history.

Good answer. Thanks for taking the time to think about it.

sthief09 · #25 10-23-2004, 05:19 PM

exactly what I was going to say. his sample size is so small when broken up into individual groups that this makes a ton of sense. I have a ~15 VP$IP now and I'm sure I regularly go through stretche when it's significantly higher or significantly lower.

Cerril · #26 10-23-2004, 06:11 PM

I'm not sure if I 'completely missed the point'... I said roughly the same thing in my post, just two up. Then again, I'm not a pro so maybe mine doesn't count. [img]/images/graemlins/grin.gif[/img]

Chris Daddy Cool · #27 10-23-2004, 06:31 PM

10k hands of 3/6. and i'm a 26/17 player with a 5.06bb/100. i must be poker god right? and it supports your theory too. why don't you play that way for 10k hands and see if you can emulate my results.

uh, sample size please.

Cosimo · #28 10-23-2004, 06:39 PM

I took a look at my stats. For the miniscule 3700 that I currently have in PT, I only looked at those where my VP$IP is greater than 20% (16.08% overall). 12 sessions (out of 67), +44.75BB, 495h, or 9 BB/100h. Ten of my top 12 winning sessions (net) had a VP$ greater than my average. My 12 tightest sessions were barely negative.

If you take a bunch of very small snapshots, some of those TAG players are going to be misclassified as LAGs, and due to their hand selection criteria their winrate will be artificially inflated -- as mine were.

Bob T. · #29 10-23-2004, 11:36 PM

Go ahead and play 25% of your hands. Let us know how that works out for you. It works for me.

ode · #30 10-24-2004, 12:44 PM

[ QUOTE ]
Go ahead and play 25% of your hands. Let us know how that works out for you. It works for me.

[/ QUOTE ]

I take it you'r BobT9 at Party.
I have to say I really like this play of yours :

Party Poker 2/4 Hold'em (10 handed)

Preflop: Hero is MP3 with T[img]/images/graemlins/heart.gif[/img], Q[img]/images/graemlins/heart.gif[/img]. CO posts a blind of $2.
UTG folds, UTG+1 (Bob T) calls, UTG+2 folds, MP1 calls, MP2 calls, Hero calls, CO (poster) checks, Button folds, SB folds, BB checks.

Flop: (6.50 SB) A[img]/images/graemlins/club.gif[/img], T[img]/images/graemlins/spade.gif[/img], 8[img]/images/graemlins/heart.gif[/img] (6 players)
BB checks, Bob T bets, MP1 raises, MP2 folds, Hero folds, CO folds, BB calls, Bob T 3-bets, MP1 caps, BB calls, Bob T calls.

Turn: (9.25 BB) Q[img]/images/graemlins/club.gif[/img] (3 players)
BB checks, Bob T bets, MP1 raises, BB calls, Bob T calls.

River: (15.25 BB) Q[img]/images/graemlins/spade.gif[/img] (3 players)
BB checks, Bob T bets, MP1 raises, BB folds, Bob T 3-bets, MP1 calls.

Final Pot: 21.25 BB
Main Pot: 21.25 BB, between Bob T and MP1. > Pot won by MP1 (21.25 BB).

Results in white below: 
Bob T has 3h 6s (one pair, queens).
MP1 has Td Tc (full house, tens full of queens).
Outcome: MP1 wins 21.25 BB. 

[img]/images/graemlins/wink.gif[/img]

/ode