Pokertracker made me do it

[MaxPower]

I've started to export player statistics from pokertracker into notes. My note on the player to my right says that his VP$IP is 27% and PFR% is 18%.

Hand 1

Party Poker 3/6 Hold'em (9 handed) converter (http://www.selachian.com/tools/bisonconverter/hhconverter.cgi)

Preflop: MaxPower is UTG+1 with J/images/graemlins/heart.gif, J/images/graemlins/diamond.gif. CO posts a blind of $4.
<font color="CC3333">UTG raises</font>, <font color="CC3333">MaxPower 3-bets</font>, MP1 folds, MP2 calls, MP3 folds, CO (poster) folds, Button folds, SB folds, BB calls, <font color="CC3333">UTG caps</font>, MaxPower calls, MP2 calls, BB calls.

Flop: (17.66 SB) 5/images/graemlins/diamond.gif, 3/images/graemlins/diamond.gif, 2/images/graemlins/heart.gif <font color="blue">(4 players)</font>
BB checks, <font color="CC3333">UTG bets</font>, <font color="CC3333">MaxPower raises</font>, MP2 calls, BB folds, UTG calls.

Turn: (11.83 BB) T/images/graemlins/heart.gif <font color="blue">(3 players)</font>
UTG checks, <font color="CC3333">MaxPower bets</font>, MP2 calls, UTG folds.

River: (13.83 BB) 4/images/graemlins/heart.gif <font color="blue">(2 players)</font>
MaxPower checks, <font color="CC3333">MP2 bets</font>, MaxPower calls.

Final Pot: 15.83 BB
<font color="#990066">Main Pot: 15.83 BB, between MaxPower and MP2.</font>

Hand 2

A few hands later

Party Poker 3/6 Hold'em (9 handed) converter (http://www.selachian.com/tools/bisonconverter/hhconverter.cgi)

Preflop: MaxPower is SB with T/images/graemlins/spade.gif, Q/images/graemlins/heart.gif.
UTG folds, UTG+1 folds, MP1 folds, MP2 folds, MP3 folds, CO folds, <font color="CC3333">Button raises</font>, <font color="CC3333">MaxPower 3-bets</font>, BB folds, Button calls.

Flop: (7 SB) K/images/graemlins/spade.gif, 2/images/graemlins/heart.gif, 7/images/graemlins/diamond.gif <font color="blue">(2 players)</font>
<font color="CC3333">MaxPower bets</font>, Button folds.

Final Pot: 4 BB
<font color="#990066">Main Pot: 3.50 BB, won by MaxPower.</font>
<font color="#990066">Pot 2: 0.50 BB, overbet by MaxPower.</font>

[cold_cash]

This might be a little off topic, but I've got a question, and it sounds like you're the man that'll have the answer.

When you import notes and statistics from PokerTracker to say, your Party Poker account, does this mean that while you're at the table you can click on a certain player, and all his stuff pops right up, just like the other player notes do?

[sthief09]

yeah, it's in advanced options. it's a really really cool feature

[J.R.]

I like hand 1, but hate hand 2, especially if you opponent is equally tancaious postflop given your positional disadvantage, limited investment ($1 sb), aggro opponent and Q high hand. Give me a decent ace or a pair and I like it more, but give up the buck here and wait for a better spot. If you play you have to 3-bet, but I don't think this is the spot.

[BugsBunny]

Given the stats on this player I like hand 2. The one stat I'd like in this position (which I export into mine) is "attempt to steal blinds". If this is high then hand 2 is a no-brainer. This is a good hand to defend with. Playing HU, or 3 handed, I play this hand about 100% of the time.

Now if the person raising from the BT has high raising standards, even when in a steal position, I might think about laying down in the SB, but would still call in the BB every time.

Against random hands QTo wins HU about 57%. 3 handed it's about 40%. Against strong raising hands (AA-99, AKs-ATs, KQs-KJs, QJs, JTs, AKo-AJo, KQo) HU it's still over 30%, and many people have looser standards then that for steal raises (any A, any pair for starters), in which case the % goes up (just adding any A and any pair to the above raising hands takes QTo to almost 40% HU).

[Gramps]

Hand 1:

Did MP1 flip over AK or AQ (probably of diamonds or hearts)? I like the 3-bet and play (postflop as well) based on your read of UTG. Kind of a "doh" when someone behind you calls 3 cold after you've 3-bet/tried to isolate with JJ, that's just a bit unlucky (assuming you're getting called with a hand that's at least semi-worthy of a 3-bet cold call).

Hand 2:

I'd be more hesitant to 3-bet here with QTo. I don't worry about protecting my blinds at full table so much (there isn't much collateral benefit to the play, how often is it going to get folded around to the button when you're in the SB at Party $3/6?). You're making it 3-bets out of position with a hand that, even against a pretty loose raiser, will probably be an underdog more times than not.

Shorthanded it's a better play, but QTo is still pushing it a bit IMO. If you've got a "only raise with a monster hand" table image then...no, a QTo 3-bet is still pushing it a bit.

[Saborion]

I believe it's important to take into consideration how many hands you have logged with the player in question, not only his stats. If those stats are after 50 hands with him, they might not be "accurate".

[MaxPower]

J.R.,

I see your point, but look at the way he played hand 1. He was clearly able to throw away a hand when played back at even when the pot is relatively large.

[MaxPower]

Bugs,

My problem with the attempting to steal blinds stat is that the opportunity to steal comes up so rarely in these games. So even if I've played a lot of hand with someone, the stat is probably based on a small number of hands. I just picked out someone who I have played 496 hands against and he has had the opportunity to steal 22 times. Next I looked at someone who I have 250 hands against and he has had the opportunity to steal 9 times.

So, unless I have a lot of hands against someone, I don't find this stat too usefull. It is more usefull to watch their play and see what kinds of hands they do try to steak with when the opporunity arises.

[BugsBunny]

In most cases I agree, but then you get someone (like I ran across yesterday) that has an attempt to steal number in the 40% range, with a semi-reasonable number of hands.

You look at that number along with his other stats. So if someone has a VPIP of 50%, with a PFR of 10% and an attempt to steal figure of 40% does that say anything to you?

[MaxPower]

Now I'm going to contradict myself within the same thread and say that sample size is not that important as long as you have 40 or more hands with the player (by the way, I picked 40 somewhat arbitrarily). Yes, if you have played 50 hands with a player his VP$IP and PFR% are not "accurate", but when do you every have accurate information in a poker game. Poker is a game of incomplete information and you often have to act on what little you know.

Let's say I'm playing live poker (or table poker - a term coined by Tommy Angelo). A players sits down and after 50 hands I notice that he is seeing half of the flops and never raising pre-flop. Am I not supposed to use this information because I have a small sample size and maybe he is just getting a lot of limping hands? Of course I should use this information.

That said, I don't refer to pokertracker stats that much, because stats don't tell the whole story. Statistics are summaries which leave out much of the important detail. The trick with statistics is knowing how to interpret them and use them to your advantage.

[MaxPower]

One other thing is that it is a little bit of a misnomer to talk about "accuracy." This assumes that there is some true value you can measure. It is not really the case because peoples playing styles change over time as they improve or get worse.

[MaxPower]

[ QUOTE ]
In most cases I agree, but then you get someone (like I ran across yesterday) that has an attempt to steal number in the 40% range, with a semi-reasonable number of hands.

You look at that number along with his other stats. So if someone has a VPIP of 50%, with a PFR of 10% and an attempt to steal figure of 40% does that say anything to you?

[/ QUOTE ]

As I said in my other reply the key to statistics is learning to interpret them and use them. So you should looked at things in their entirety and make some assumptions based upon them.

I'm still a little wary of assuming too much based upon something like 9 observations. Perhaps you play much more than me, but the vast majority of players in my database have less than 200 hands. Most of the players I have a large number of hands against against are relatively tight-aggressive.

[BugsBunny]

I've noticed that myself (most players with a large number of hands are tight-aggressive). Wonder why that is? Hmm - maybe they'e making money while the others are losing?

Yes, I agree with everything you say. Stats have to be looked at in context and not in a vacuum. They're simply another tool in the arsenal, and not even close to being the most important one. The more I play the less I depend on Pokertracker during a game. But there are times it comes in very handy.

Another example. I'm in a game with AQo and someone raises in front of me. I take a quick look and he has about 60 hands in there with a PFR of about 1.5%. Guess what I do with that AQ? Don't care how many other people are in the pot.

But back to this case. Judging something on 9 hands is **very** dangerous, although I would tend to think that a 44% or greater stat still indicates some sort of probability. If someone has attempted to steal 4 out of 9 times it's *likely* that his steal standards are loose (although not impossible that he just had good hands). If that number was 7 out of 9 I'd be pretty certain of it.

If you just sat down at a table and someone raised preflop with 4 of his first 9 hands would you just ignore that because the sample size was so small? I think you would tentatively label him as a loose raiser, with a potential to be maniacal - pending further developments.

[J.R.]

If that's the case, its not too bad, perhaps a little marginal against many but profitable against weak-tight postflop players. That's why I was curious whether you opponent was equally tenacious postflop. If hand 1 is indicative of his play, OK, but I tend to find people who are very if not hyper aggresive preflop (raising 18%) continue to be so postflop, especially with position and a limited field of opponents.

But if opponent is as weak postflop as you seem to imply, that's a goldmine- he builds shorthanded (unless people are coldcalling his raise) pots to give to other players- the aggressive ones.