[OFFICIAL] Stats and Graphs Analysis Thread
03-18-2008
, 10:49 PM
The purpose of this thread is to provide some analysis of the Poker Tracker stats that new players post when they ask for a check-up. If you want to be playing a tight/aggressive style, and if your stats deviate from the stats analyzed in this thread, it is probably for one of the reasons described. Do the legwork in poker tracker, analyze your hand histories, and post specific hands that might be symptomatic of your leaks. If you read this, there should not be any reason for you to post a stat check-up; you should have more specific questions that are better addressed as hand history posts.
This is the classic 2+2 post on Poker Tracker stats
INTRO
Your Poker Tracker stats do not matter in and of themselves. Rather, they are describing the way you play. If you learn to listen to what your Poker Tracker stats are saying, and adjust appropriately, you will play better, up to a point.
Here are some stats from two good Tags at NL $50 and $100:
Baja 15's stats
chargers in 07’s stats
As you can see, their stats are different, but they share certain characteristics that show that they are playing solid, tight/aggressive poker. Here is what chargers’ and baja’s stats are saying:
1. Seizing initiative:
baja15 is winning at 17.5/13.5. chargers in o7 is winning at 13.4/11. The specific numbers are not important; what is important is that both chargers and baja are almost always raising the pots they play. Seizing the initiative is crucial to winning at tight/aggressive poker. They have the table reacting to them. When they call, which is usually about 3.5% of the pots, they have a reason for doing so. If chargers or baja went back through all the hands they called a PFR with, they would be able to tell you why they did it every time. And they would never say “well, because KQ is playable in MP” or something vague like that. It would be “because I was getting the implied odds to play 33 for a set here—villain is a 10/2 nit with 100bb who only raises UTG with premium hands; I knew I’d get at least a big part of his stack if I hit my set.”
The bottom line here is that by raising when you enter a pot, you seize the initiative. If you are not raising about 2/3 of the hands you play, you are playing too passively.
2. Positional awareness:
The closer they are to the button, the more hands chargers and baja play. They understand that a hand that is junk in EP (ATo) becomes a raising hand when it is folded to them in LP. This is applied probability—the chance that someone behind you has a better hand increases when there are more players behind you. Say you are first to act of 9 players. There are 8 behind you. If you have Q7s (the hand that is the median hand in NLHE), on average 4 players behind you will have a better hand and 4 will have worse. If you are UTG with ATo, a top 1/3-ish hand, you can expect about 2-3 players to have a better hand. So you fold it. That same hand, OTB, when it has been folded to you, rates to be the best of the 3 remaining hands. So you raise. In EP Tags usually raise with premium hands plus a few hands that fit their preferred style of play (all pocket pairs, suited or connected paint) and fold everything else. They loosen their starting requirements progressively to the point that in the CO and OTB the fact that they are in position is as important as the cards they happen to have been dealt.
Notice how baja and chargers’ cold called a PFR% increases as they approach the button. They become more willing to call a preflop raise as their position improves, because they know they can win some hands by using position even when they don’t make a hand. Out of position, they are ruthless in applying raise or fold to their hand selection. OOP they have to take the initiative; in position, they look for spots to use position to use another player’s aggression against him.
If you do not have a similar ratio, you should be playing fewer hands UTG and more OTB.
3. Blind stealing.
This subject blends into positional awareness. In addition to raising for value with decent hands that you expect to be the favorite, you should raise with some hands you would ordinarily fold, hoping that it gets folded around. For an excellent discussion of blind stealing, carefully study and apply Pokey's and Dan Bitel’s advice on blind stealing. chargers has achieved what Pokey thinks is the optimal 30% steal percentage, and Baja is right there, too. Pokey does the math on how profitable this can be, and baja and chargers sure look like solid evidence Pokey is right. Pokey also talks about what hands are good stealing hands.
The combination of raising for value in LP with hands that you fold in EP and raising with folding hands to steal the blinds should get your CO and OTB VPIP up above 20%, and closer to 30% is probably best for experienced players.
4. Postflop Aggression.
chargers’ AF is 4.48. PT calculates your AF by adding the number of times you bet to the number of times you raise, and dividing this sum by the number of times you call. chargers bets or raises 4.5 times as often as he calls.
Your total AF doesn’t tell the whole story, though. Look at chargers’ aggression by street—he is way more aggressive on the flop (6.3) than he is on the river (2.29). This high AF means:
He decides on the flop whether and how he will play the hand. One way you increase your aggression factor is to fold to flop bets. chargers probably never says to himself “hmm, ok, I’ll call and we’ll see what he does on the turn.” He puts his opponent on a range preflop based on his opponent’s action. Then he looks at what his opponent does on the flop, and adjusts the range. If he decides to play, he bets or raises six times as often as he calls on the flop, and over twice as often on later streets. He can be this aggressive because he folds on the flop when he thinks the flop hit his opponent’s range pretty hard while missing him.
chargers does not slow play often. Slow playing trashes your aggression factor. Chargers has internalized the correct strategy of betting out or raising with strong hands most of the time. If your aggression factor is under 3, go back and study your monsters and check to see how often you are slow playing them. You shouldn’t be, very often, and you should always have a very player or board specific reason for doing so. If you can’t list those reasons off the top of your head right now, you should search the forum and learn when it is appropriate to slow play. On the other hand, you should be calling so few bets that slow playing monsters represents a significant fraction of your total number of calls.
If chargers plays a draw, he plays it aggressively. Check/calling oop or calling in position when you have a draw will drag down your aggression factor. Again, we don’t care about the aggression factor per se, but we do care that it may be telling us that we are not playing draws aggressively. Betting or raising with very good draws gives you two ways to win the pot, but just calling with them means you are playing fit or fold—you have to make your hand to have a chance of winning the pot. By betting out with his very good draws a lot of the time he decides to play them, chargers is winning a lot of them without ever making his draw. By folding his decent draws in the face of big bets and lots of aggression, he is saving money on draws that are too marginal to bet or raise with against a determined opponent, and these folds will also increase his aggression factor.
The lower your aggression factor, the more likely it is that you are making the common mistakes that Chargers does not make very often (calling to see what happens next or because you are unsure of where you are, slow playing too much, and not playing draws aggressively). Read this, too
5. WTSD%, W$WSF, W$SD: These stats come the closest of all stats to having hard and fast rules associated with them. As a general rule, winning players usually:
Go to showdown about 25% of the time
Win when they see the flop at least 35% of the time
Win at showdown between 45 and 48% of the time OR between 52 and 55% of the time.
WTSD%:
If this stat is too far from 25%, it could indicate several different problems. You could either be not betting enough on later streets when you are ahead, or you could be calling too much behind. Usually, when this stat is off, it is both. Your by street aggression factor is a huge tell here. If you are below 2 on any street, you are going for pot control too much (usually on the turn and the river). Make your read on the flop—if you are behind with little chance to improve, fold on the flop. If you think you are ahead or have a strong draw—bet or raise a lot, and win the hand before the river (if you are actually behind, your opponent WILL let you know, normally by sliding it far to the right and clicking). Either way, it is a hand that doesn’t go to showdown. If you think your opponent has a decent hand, fold or raise—either get out or try to push him off his marginal hand. Either way it is a hand that doesn’t go to showdown.
The correct action on the river is a very complicated subject, but, in general, if you are checking behind a lot in position, you are increasing your WSD% by failing to bet and induce a fold from an opponent who checked to you. Only check behind when there are very few or no hands that you beat that would call a bet. If your WSD% is too high, you need to use Poker Tracker and Poker EV to look at the hands where you did not bet the river and won. Then, in the future, you bet in similar situations. Your opponents will fold a lot, and you will decrease your WSD%. Then you need to find hands where you called on the river and lost. Among these hands, you will find hands where it was correct to have folded on an earlier street, maybe even preflop. Do so in future similar situations, and you will decrease your WSD% and increase your winrate by winning more when you are good, and losing less when you are behind.
If you make these adjustments, you will be playing better, and, incidentally, you will see the change reflected in your WSD%
W$WSF%.
If you are not winning at least 35% of the hands you see the flop, there are 3 possible problems, all of which will need examination:
You are playing marginal hands too much. 97s or 22 are great hands to speculate with in ideal situations (deep, multiway and in position), but win like 12% of the time. If you play these hands indiscriminately, it will show up as a lower W$WSF%
You are getting blown off hands by aggression. Remember, most flops miss most players.
You are not winning pots based on position and you are not c-betting enough.
All 3 of these are symptomatic of “fit or fold,” poker, which is –EV poker.
W$atSD
There are two approaches Tags take to showdown. Both are profitable, but they show up as different ranges of winning at SD:
Tags who border on loose aggressive play generally win less than 50% of their showdowns. This reflects the fact that they are pushing very hard, often all in, with good draws or marginal hands, relying on their opponents’ willingness to fold as much as the quality of their hands to win pots. They frequently make big bets when they know they are behind but have outs, hoping for a fold, but knowing they have plenty of cards as outs if they get called. Here’s an example of a Lag sample from ship_it_trebek that shows you can achieve a nice winrate even when you lose most showdowns:
If you play a typical Tag style, you should be above 50%. If you are not, you have to go back through and look for patterns in your river decisions to find the recurring mistakes (there WILL be recurring mistakes, most likely calling a big river bet with TPGK or an overpair). If your W$atSD is too high (yes, it is possible) it means that you almost have to be folding the best hand too often.
Conclusion
I am up against the max word count and there’s a ton I had to cut to get this much done. I’ll ask the experienced players to add their thoughts on some of the things I didn’t get to, such as defending blinds, and to point out any mistakes I made.
Part 2, which I will post as a reply, will be a practical exercise I think the newer players should participate in.
This is the classic 2+2 post on Poker Tracker stats
INTRO
Your Poker Tracker stats do not matter in and of themselves. Rather, they are describing the way you play. If you learn to listen to what your Poker Tracker stats are saying, and adjust appropriately, you will play better, up to a point.
Here are some stats from two good Tags at NL $50 and $100:
Baja 15's stats
chargers in 07’s stats
As you can see, their stats are different, but they share certain characteristics that show that they are playing solid, tight/aggressive poker. Here is what chargers’ and baja’s stats are saying:
1. Seizing initiative:
baja15 is winning at 17.5/13.5. chargers in o7 is winning at 13.4/11. The specific numbers are not important; what is important is that both chargers and baja are almost always raising the pots they play. Seizing the initiative is crucial to winning at tight/aggressive poker. They have the table reacting to them. When they call, which is usually about 3.5% of the pots, they have a reason for doing so. If chargers or baja went back through all the hands they called a PFR with, they would be able to tell you why they did it every time. And they would never say “well, because KQ is playable in MP” or something vague like that. It would be “because I was getting the implied odds to play 33 for a set here—villain is a 10/2 nit with 100bb who only raises UTG with premium hands; I knew I’d get at least a big part of his stack if I hit my set.”
The bottom line here is that by raising when you enter a pot, you seize the initiative. If you are not raising about 2/3 of the hands you play, you are playing too passively.
2. Positional awareness:
The closer they are to the button, the more hands chargers and baja play. They understand that a hand that is junk in EP (ATo) becomes a raising hand when it is folded to them in LP. This is applied probability—the chance that someone behind you has a better hand increases when there are more players behind you. Say you are first to act of 9 players. There are 8 behind you. If you have Q7s (the hand that is the median hand in NLHE), on average 4 players behind you will have a better hand and 4 will have worse. If you are UTG with ATo, a top 1/3-ish hand, you can expect about 2-3 players to have a better hand. So you fold it. That same hand, OTB, when it has been folded to you, rates to be the best of the 3 remaining hands. So you raise. In EP Tags usually raise with premium hands plus a few hands that fit their preferred style of play (all pocket pairs, suited or connected paint) and fold everything else. They loosen their starting requirements progressively to the point that in the CO and OTB the fact that they are in position is as important as the cards they happen to have been dealt.
Notice how baja and chargers’ cold called a PFR% increases as they approach the button. They become more willing to call a preflop raise as their position improves, because they know they can win some hands by using position even when they don’t make a hand. Out of position, they are ruthless in applying raise or fold to their hand selection. OOP they have to take the initiative; in position, they look for spots to use position to use another player’s aggression against him.
If you do not have a similar ratio, you should be playing fewer hands UTG and more OTB.
3. Blind stealing.
This subject blends into positional awareness. In addition to raising for value with decent hands that you expect to be the favorite, you should raise with some hands you would ordinarily fold, hoping that it gets folded around. For an excellent discussion of blind stealing, carefully study and apply Pokey's and Dan Bitel’s advice on blind stealing. chargers has achieved what Pokey thinks is the optimal 30% steal percentage, and Baja is right there, too. Pokey does the math on how profitable this can be, and baja and chargers sure look like solid evidence Pokey is right. Pokey also talks about what hands are good stealing hands.
The combination of raising for value in LP with hands that you fold in EP and raising with folding hands to steal the blinds should get your CO and OTB VPIP up above 20%, and closer to 30% is probably best for experienced players.
4. Postflop Aggression.
chargers’ AF is 4.48. PT calculates your AF by adding the number of times you bet to the number of times you raise, and dividing this sum by the number of times you call. chargers bets or raises 4.5 times as often as he calls.
Your total AF doesn’t tell the whole story, though. Look at chargers’ aggression by street—he is way more aggressive on the flop (6.3) than he is on the river (2.29). This high AF means:
He decides on the flop whether and how he will play the hand. One way you increase your aggression factor is to fold to flop bets. chargers probably never says to himself “hmm, ok, I’ll call and we’ll see what he does on the turn.” He puts his opponent on a range preflop based on his opponent’s action. Then he looks at what his opponent does on the flop, and adjusts the range. If he decides to play, he bets or raises six times as often as he calls on the flop, and over twice as often on later streets. He can be this aggressive because he folds on the flop when he thinks the flop hit his opponent’s range pretty hard while missing him.chargers does not slow play often. Slow playing trashes your aggression factor. Chargers has internalized the correct strategy of betting out or raising with strong hands most of the time. If your aggression factor is under 3, go back and study your monsters and check to see how often you are slow playing them. You shouldn’t be, very often, and you should always have a very player or board specific reason for doing so. If you can’t list those reasons off the top of your head right now, you should search the forum and learn when it is appropriate to slow play. On the other hand, you should be calling so few bets that slow playing monsters represents a significant fraction of your total number of calls.If chargers plays a draw, he plays it aggressively. Check/calling oop or calling in position when you have a draw will drag down your aggression factor. Again, we don’t care about the aggression factor per se, but we do care that it may be telling us that we are not playing draws aggressively. Betting or raising with very good draws gives you two ways to win the pot, but just calling with them means you are playing fit or fold—you have to make your hand to have a chance of winning the pot. By betting out with his very good draws a lot of the time he decides to play them, chargers is winning a lot of them without ever making his draw. By folding his decent draws in the face of big bets and lots of aggression, he is saving money on draws that are too marginal to bet or raise with against a determined opponent, and these folds will also increase his aggression factor.The lower your aggression factor, the more likely it is that you are making the common mistakes that Chargers does not make very often (calling to see what happens next or because you are unsure of where you are, slow playing too much, and not playing draws aggressively). Read this, too
5. WTSD%, W$WSF, W$SD: These stats come the closest of all stats to having hard and fast rules associated with them. As a general rule, winning players usually:
Go to showdown about 25% of the timeWin when they see the flop at least 35% of the timeWin at showdown between 45 and 48% of the time OR between 52 and 55% of the time.WTSD%:
If this stat is too far from 25%, it could indicate several different problems. You could either be not betting enough on later streets when you are ahead, or you could be calling too much behind. Usually, when this stat is off, it is both. Your by street aggression factor is a huge tell here. If you are below 2 on any street, you are going for pot control too much (usually on the turn and the river). Make your read on the flop—if you are behind with little chance to improve, fold on the flop. If you think you are ahead or have a strong draw—bet or raise a lot, and win the hand before the river (if you are actually behind, your opponent WILL let you know, normally by sliding it far to the right and clicking). Either way, it is a hand that doesn’t go to showdown. If you think your opponent has a decent hand, fold or raise—either get out or try to push him off his marginal hand. Either way it is a hand that doesn’t go to showdown.
The correct action on the river is a very complicated subject, but, in general, if you are checking behind a lot in position, you are increasing your WSD% by failing to bet and induce a fold from an opponent who checked to you. Only check behind when there are very few or no hands that you beat that would call a bet. If your WSD% is too high, you need to use Poker Tracker and Poker EV to look at the hands where you did not bet the river and won. Then, in the future, you bet in similar situations. Your opponents will fold a lot, and you will decrease your WSD%. Then you need to find hands where you called on the river and lost. Among these hands, you will find hands where it was correct to have folded on an earlier street, maybe even preflop. Do so in future similar situations, and you will decrease your WSD% and increase your winrate by winning more when you are good, and losing less when you are behind.
If you make these adjustments, you will be playing better, and, incidentally, you will see the change reflected in your WSD%
W$WSF%.
If you are not winning at least 35% of the hands you see the flop, there are 3 possible problems, all of which will need examination:
You are playing marginal hands too much. 97s or 22 are great hands to speculate with in ideal situations (deep, multiway and in position), but win like 12% of the time. If you play these hands indiscriminately, it will show up as a lower W$WSF%You are getting blown off hands by aggression. Remember, most flops miss most players. You are not winning pots based on position and you are not c-betting enough.All 3 of these are symptomatic of “fit or fold,” poker, which is –EV poker.
W$atSD
There are two approaches Tags take to showdown. Both are profitable, but they show up as different ranges of winning at SD:
Tags who border on loose aggressive play generally win less than 50% of their showdowns. This reflects the fact that they are pushing very hard, often all in, with good draws or marginal hands, relying on their opponents’ willingness to fold as much as the quality of their hands to win pots. They frequently make big bets when they know they are behind but have outs, hoping for a fold, but knowing they have plenty of cards as outs if they get called. Here’s an example of a Lag sample from ship_it_trebek that shows you can achieve a nice winrate even when you lose most showdowns:If you play a typical Tag style, you should be above 50%. If you are not, you have to go back through and look for patterns in your river decisions to find the recurring mistakes (there WILL be recurring mistakes, most likely calling a big river bet with TPGK or an overpair). If your W$atSD is too high (yes, it is possible) it means that you almost have to be folding the best hand too often.Conclusion
I am up against the max word count and there’s a ton I had to cut to get this much done. I’ll ask the experienced players to add their thoughts on some of the things I didn’t get to, such as defending blinds, and to point out any mistakes I made.
Part 2, which I will post as a reply, will be a practical exercise I think the newer players should participate in.
Last edited by Cry Me A River; 03-19-2008 at 12:53 PM.
03-18-2008
, 10:49 PM
OK, this is part 2, the quiz! For new players who need work on interpreting PT stats, here’s the background:
Four months ago I was a dyed in the wool nit. Here are my stats from September 1, 2007 to December 31, 2007:
These stats show that my 2007 game had some big problems. Go through my old stats, and point out in a reply to this thread what you think were the problems with the way I was playing. Be specific. Don’t just say “your VPIP was too high/too low.” Look at the details and tell me how I should have been playing my hands differently. There are 8 very obvious leaks in my 2007 game that are easily inferred from reviewing my PT stats (yes, I sucked).
Here are my new stats; still not great, but I think we can all agree it is a significant improvement:
Again, as an exercise in PT stat interpretation, explain specifically how I changed the way I played. Anybody who is interested can feel free to identify the areas in which I need to continue to improve, as well. I am pretty sure I am aware of all of them, but it doesn’t hurt to have second opinions.
Four months ago I was a dyed in the wool nit. Here are my stats from September 1, 2007 to December 31, 2007:
These stats show that my 2007 game had some big problems. Go through my old stats, and point out in a reply to this thread what you think were the problems with the way I was playing. Be specific. Don’t just say “your VPIP was too high/too low.” Look at the details and tell me how I should have been playing my hands differently. There are 8 very obvious leaks in my 2007 game that are easily inferred from reviewing my PT stats (yes, I sucked).
Here are my new stats; still not great, but I think we can all agree it is a significant improvement:
Again, as an exercise in PT stat interpretation, explain specifically how I changed the way I played. Anybody who is interested can feel free to identify the areas in which I need to continue to improve, as well. I am pretty sure I am aware of all of them, but it doesn’t hurt to have second opinions.
03-18-2008
, 11:03 PM
Fixed, I think. They were chargers in 07's stats.
sorry to both of you.
any remaining references to threads' stats are a mistake, they should say chargers.
Also: I got lots of stats from awesome players, threads included. I sort of picked chargers and bajas because they were the first ones I saw, no other reason.
Thanks to everybody who made their stats available, whether I used them, didn't, or thought I did but didn't!
sorry to both of you.
any remaining references to threads' stats are a mistake, they should say chargers.
Also: I got lots of stats from awesome players, threads included. I sort of picked chargers and bajas because they were the first ones I saw, no other reason.
Thanks to everybody who made their stats available, whether I used them, didn't, or thought I did but didn't!
03-18-2008
, 11:23 PM
Quote:
Fixed, I think. They were chargers in 07's stats.
sorry to both of you.
any remaining references to threads' stats are a mistake, they should say chargers.
Also: I got lots of stats from awesome players, threads included. I sort of picked chargers and bajas because they were the first ones I saw, no other reason.
Thanks to everybody who made their stats available, whether I used them, didn't, or thought I did but didn't!
sorry to both of you.
any remaining references to threads' stats are a mistake, they should say chargers.
Also: I got lots of stats from awesome players, threads included. I sort of picked chargers and bajas because they were the first ones I saw, no other reason.
Thanks to everybody who made their stats available, whether I used them, didn't, or thought I did but didn't!
03-19-2008
, 01:14 AM
03-19-2008
, 01:14 AM
1. I went back through and looked at a lot of stat check up posts and, iirc, all of the winning players were in the neighborhood of 25% some were a bit higher (26, 27), some were lower (20-23). I don't recall ever seeing a winning player who wasn't in the 20s somewhere. Small sample size, maybe 30 stat check ups, but telling, imo.
2. 30% is the threshhold that gets you autorated as a calling station by the excellent replacement PT rules that are linked elsewhere on 2+2, but that I can't find at the moment.
FWIW, I would be inclined to prefer 20% to 30%.
If you are at 20%, imo, there's a good chance you are folding the best hand on a regular, but fairly infrequent, basis. But it is also possible that you are blowing opponents off of hands too early with big bets on earlier streets. I'd say look at your WSD% in connection with your by street AF and attempt to determine which is more likely, then look through your hand histories to see if you can find empirical support for the hypothesis that looks more likely.
Edit: I left in a bunch of references to threads13. Can a mod change them to read "chargers"? It is irritating the bejeepers out of me for this post to be so sloppy.
Last edited by mpethybridge; 03-19-2008 at 01:24 AM.
Reason: Cuz I suck at editing and my time has expired.
03-19-2008
, 02:14 AM
03-19-2008
, 02:19 AM
Very nice post, sticky this!!
03-19-2008
, 02:52 AM
I didn't look too much at the post, but it seemed nicely put together. Also, chargers now has a higher pfr than he had vpip a while back
.
03-19-2008
, 02:57 AM
03-19-2008
, 03:00 AM
03-19-2008
, 03:23 AM
Quote:
and in general, yeah, see, all the +30% WTSD and <20%WTSD winners are all coming out of the woodwork now!!! Where were all of you when I was thinking I had discovered a solid guideline?
Crap--I hate it when an inconvenient fact gets in the way of a pretty theory.
I still think 25% is a solid goal for new players, but it also seems that some people win with WTSD% outside of the 20s. It may be that 25% is still closer to optimal, whatever that means at uFR.
And I definitely believe that a new player outside the mid-20s is a recipe for disaster.