Limit Lessons--Aggression and Winning
A lot of people on Bet the Pot ask in the LHE Forum about how they can improve their game and win more consistently. Then they post their stats, and invariably, their aggresion stats look like this:
Flop 27.19 raise+bet / 15.44 call = 1.76
Turn 40.66 raise+bet / 21.64 call = 1.88
River 35.23 raise+bet/ 26.42 call = 1.33
When I did the research to put together my auto-rate rules for PT, I discovered that your pre-flop hand selection doesn't really have anything to do with your winrate. Pre-flop in LHE is much like swinging the driver in golf. Some players who are bombers off the tee (like myself) can't break 90, while other players who hit shorter, but consistently shoot in the 70's. In golf, you drive for show and putt for dough. In Holdem, you play pre-flop for show, but your money is made entirely on how well you play after the flop hits the table. That is where aggression comes into play.
My research found the following (from my 1.5 million+ observed hand database):
Rating Category................................................BB/100
Ultra Aggressor (AF-T>2.00).........................3.59
Tight Aggressive (AF-T 1.5-2).......................1.98
Semi-Loose Aggressive...........................................1.97
Loose Aggressive................................................0.72
Tight Neutral (AF-T 1.2-1.5)........................0.45
Semi-Loose Neutral..............................................1.79
Loose Neutral...................................................0.71
Tight Passive (AF-T<1.2)...........................-3.70
Semi-Loose Passive.............................................-11.42
Loose Passive...................................................-14.41
These numbers are not by coincidence. Your post-flop play is by far the primary dictating factor in your ability to beat this game. Without aggression, you simply will NOT WIN LONG TERM. I had made the following assertion in the forum:
I actually got some negative feedback:
I almost fell on the floor. Jennifer Harman, one of the best LHE players in the world, in her section in SuperSystem 2, basically advocates ruthless aggression in exactly the way I mentioned--bet your strong hands, bet your strong draws. I play a tighter style, so when I'm in a pot, especially when I've raised pre-flop, I'm firing away, because to me, my hand is best until something tells me (board, another player's action) tells me it's not. In LHE, you can't bet someone out of a pot, so the only way you can protect a hand (see previous post) is to bet-bet-bet it until the other player folds or you win at showdown.
The research, and not just my own, either, states clearly that without aggression, you will not win! You have very little chance to actually protect a hand, so playing aggressively is the only way you can consistently win in this game.
Consider this example. You're in MP, looking at AK. The player to your right limps, you raise, and the button, BB, and the limper call your raise. The flop comes a Jack-high rainbow, and the limper checks. What do you do here? Bet. No doubt about it. You have overcards, and unless you think you're drawing behind 2 pair, a set, or a straight or flush draw (and your reads should give you an idea here), you need to bet. The only information you've given the opposition is that you have a strong hand pre-flop--you've raised. Maintain the strength. A bet tells them that you're not afraid of the Jack, or any other card on that flop. Many players will put you on QQ, KK, or AA and dump the hand. Others will call the bet and potentially give you a free card on the turn/river. This is clearly +EV.
Change the example slightly. The pre-flop action is the same, only the limper bets. What do you do now? You RAISE. Why? You will likely limit the field and make it a heads-up situation on the turn and river. Also, you are basically tossing in a feeler to see how serious he is about continuing with this hand. If he has Jack-rag, he's gotta think (if he's not a calling station) that he's beat here 8 ways to Sunday, and that the WORST you have is AJ, while you could have a set, QQ, KK, or AA. The only situations you're worrying about (and this is again where reads come into play) is if he has a hand like KJ, since that essentially nullifies 3 of your outs. If you're sure he doesn't have KJ, AJ, or a set or 2 pair (in which case you're screwed), this play works fantastically. You have a minimum of 6 good outs and you have him playing defense the rest of the way. He goes passive, and while remembering that you can't make him fold if you don't bet, you now have the mental edge in the hand to catch your outs, make him fold, or perhaps even win with your AK unimproved.
This is how aggression works in your favor. In the above example, if he has middle or bottom pair, your raise tells him that his middle/bottom pair just CAN'T be good. Most times, he'll check the turn, and your bet will many times force him to fold. The only issue is if he has a top pair hand or better. When combined with the fact that most flops miss most hands, it is fairly clear that this strategy has a positive expectation.
Here's the caveat--Don't ever try to semi-bluff/bluff the calling stations. You will lose if you don't catch. Frequently. This is where PokerTracker and PokerAce HUD come into play. If you're not playing with these tools on your side, you're losing money. Period. Much of poker is reading people, reading hands. Online, it's much harder to do this, so programs like PAHud and PT get you the information you'd have if you were playing live.
If you're still not believing that the above strategy has a positive expectation, consider this:
You have AK, the limper has JT. The flop is J-7-3 rainbow. There is no doubt that you're at least temporarily behind here, right? There are 13 SB in the pot on the flop, and the limper leads out for a bet. With 14 SB in the pot, you can call with 14-to-1 odds to see a turn card. If the turn card is a blank (or otherwise doesn't help you), you can dump it, right?
Wrong. Same idea. With 14 SB in the pot, you RAISE. Players downstream of you now have to call 2 bets to win 14 (7-to-1 odds). If they don't have a top pair hand or better, calling for them is a mistake. They can figure that they're behind at least one hand (the original limper/bettor), and most likely TWO hands, since the first bet didn't even faze you and you raised. They both fold, bringing the action back to the limper/bettor. He called a pre-flop raise, and he has just been raised by a strong hand (indicated by the PFR) after he led. He's getting 16-to-1 to call here with his top pair, mediocre kicker hand, and he's figuring that now, he's behind, perhaps severely behind. He calls, and the turn brings:
Example 1: A jack, ten, or the board pairs (7 or 3). He leads out for 1 BB, and you figure you have to be behind. If your read is that he's bluffing, you call and hope to hit on the river, or you can fold--net loss 4 SB.
Example 2: An ace or king. He checks, you bet, he folds. You win 9 BB.
Example 2a: An ace or king. He checks, you bet, he calls, river a blank. Check-bet-fold. You win 10 BB.
Example 2b: An ace or king. He checks, you bet, he calls, river a blank. Check-bet-call. You win 11 BB.
Example 3: Anything else. He checks, you bet. If he doesn't improve, he almost has to fold, unless he puts you on the unimproved hand you have in front of you.
See where this helps? Even checking behind on the turn has value in that it gives you a free river card and allows you another chance to catch up. An ace or king may fall on the river, he may actually bet into you and you can raise and win a nice pot, or another blank can fall and you can drop the hand for little loss.
I hope this helps explain the importance of aggression in LHE, and if you take nothing else from this, take these two things:
Flop 27.19 raise+bet / 15.44 call = 1.76
Turn 40.66 raise+bet / 21.64 call = 1.88
River 35.23 raise+bet/ 26.42 call = 1.33
When I did the research to put together my auto-rate rules for PT, I discovered that your pre-flop hand selection doesn't really have anything to do with your winrate. Pre-flop in LHE is much like swinging the driver in golf. Some players who are bombers off the tee (like myself) can't break 90, while other players who hit shorter, but consistently shoot in the 70's. In golf, you drive for show and putt for dough. In Holdem, you play pre-flop for show, but your money is made entirely on how well you play after the flop hits the table. That is where aggression comes into play.
My research found the following (from my 1.5 million+ observed hand database):
Rating Category................................................BB/100
Ultra Aggressor (AF-T>2.00).........................3.59
Tight Aggressive (AF-T 1.5-2).......................1.98
Semi-Loose Aggressive...........................................1.97
Loose Aggressive................................................0.72
Tight Neutral (AF-T 1.2-1.5)........................0.45
Semi-Loose Neutral..............................................1.79
Loose Neutral...................................................0.71
Tight Passive (AF-T<1.2)...........................-3.70
Semi-Loose Passive.............................................-11.42
Loose Passive...................................................-14.41
These numbers are not by coincidence. Your post-flop play is by far the primary dictating factor in your ability to beat this game. Without aggression, you simply will NOT WIN LONG TERM. I had made the following assertion in the forum:
As in my autorate rule discussion in another stickied thread, aggression is the key to winning poker. Plain and simple. No doubt about it. Pre-flop aggression factor is absolutely worthless. Ignore it. Your VP$IP and PFR aren't bad, but if you play passive post-flop, you will lose. Raise with your made hands, raise with your draws. Raise more than call, Fold more than raise.
I actually got some negative feedback:
...but this sounds more like blind aggression than selective aggression. I can't imagine what would happen to my standard deviation (15BB/100hands) if I jammed even harder.
Any thoughts on standard deviation and where it outght to be in online Limit games?
I almost fell on the floor. Jennifer Harman, one of the best LHE players in the world, in her section in SuperSystem 2, basically advocates ruthless aggression in exactly the way I mentioned--bet your strong hands, bet your strong draws. I play a tighter style, so when I'm in a pot, especially when I've raised pre-flop, I'm firing away, because to me, my hand is best until something tells me (board, another player's action) tells me it's not. In LHE, you can't bet someone out of a pot, so the only way you can protect a hand (see previous post) is to bet-bet-bet it until the other player folds or you win at showdown.
The research, and not just my own, either, states clearly that without aggression, you will not win! You have very little chance to actually protect a hand, so playing aggressively is the only way you can consistently win in this game.
Consider this example. You're in MP, looking at AK. The player to your right limps, you raise, and the button, BB, and the limper call your raise. The flop comes a Jack-high rainbow, and the limper checks. What do you do here? Bet. No doubt about it. You have overcards, and unless you think you're drawing behind 2 pair, a set, or a straight or flush draw (and your reads should give you an idea here), you need to bet. The only information you've given the opposition is that you have a strong hand pre-flop--you've raised. Maintain the strength. A bet tells them that you're not afraid of the Jack, or any other card on that flop. Many players will put you on QQ, KK, or AA and dump the hand. Others will call the bet and potentially give you a free card on the turn/river. This is clearly +EV.
Change the example slightly. The pre-flop action is the same, only the limper bets. What do you do now? You RAISE. Why? You will likely limit the field and make it a heads-up situation on the turn and river. Also, you are basically tossing in a feeler to see how serious he is about continuing with this hand. If he has Jack-rag, he's gotta think (if he's not a calling station) that he's beat here 8 ways to Sunday, and that the WORST you have is AJ, while you could have a set, QQ, KK, or AA. The only situations you're worrying about (and this is again where reads come into play) is if he has a hand like KJ, since that essentially nullifies 3 of your outs. If you're sure he doesn't have KJ, AJ, or a set or 2 pair (in which case you're screwed), this play works fantastically. You have a minimum of 6 good outs and you have him playing defense the rest of the way. He goes passive, and while remembering that you can't make him fold if you don't bet, you now have the mental edge in the hand to catch your outs, make him fold, or perhaps even win with your AK unimproved.
This is how aggression works in your favor. In the above example, if he has middle or bottom pair, your raise tells him that his middle/bottom pair just CAN'T be good. Most times, he'll check the turn, and your bet will many times force him to fold. The only issue is if he has a top pair hand or better. When combined with the fact that most flops miss most hands, it is fairly clear that this strategy has a positive expectation.
Here's the caveat--Don't ever try to semi-bluff/bluff the calling stations. You will lose if you don't catch. Frequently. This is where PokerTracker and PokerAce HUD come into play. If you're not playing with these tools on your side, you're losing money. Period. Much of poker is reading people, reading hands. Online, it's much harder to do this, so programs like PAHud and PT get you the information you'd have if you were playing live.
If you're still not believing that the above strategy has a positive expectation, consider this:
You have AK, the limper has JT. The flop is J-7-3 rainbow. There is no doubt that you're at least temporarily behind here, right? There are 13 SB in the pot on the flop, and the limper leads out for a bet. With 14 SB in the pot, you can call with 14-to-1 odds to see a turn card. If the turn card is a blank (or otherwise doesn't help you), you can dump it, right?
Wrong. Same idea. With 14 SB in the pot, you RAISE. Players downstream of you now have to call 2 bets to win 14 (7-to-1 odds). If they don't have a top pair hand or better, calling for them is a mistake. They can figure that they're behind at least one hand (the original limper/bettor), and most likely TWO hands, since the first bet didn't even faze you and you raised. They both fold, bringing the action back to the limper/bettor. He called a pre-flop raise, and he has just been raised by a strong hand (indicated by the PFR) after he led. He's getting 16-to-1 to call here with his top pair, mediocre kicker hand, and he's figuring that now, he's behind, perhaps severely behind. He calls, and the turn brings:
Example 1: A jack, ten, or the board pairs (7 or 3). He leads out for 1 BB, and you figure you have to be behind. If your read is that he's bluffing, you call and hope to hit on the river, or you can fold--net loss 4 SB.
Example 2: An ace or king. He checks, you bet, he folds. You win 9 BB.
Example 2a: An ace or king. He checks, you bet, he calls, river a blank. Check-bet-fold. You win 10 BB.
Example 2b: An ace or king. He checks, you bet, he calls, river a blank. Check-bet-call. You win 11 BB.
Example 3: Anything else. He checks, you bet. If he doesn't improve, he almost has to fold, unless he puts you on the unimproved hand you have in front of you.
See where this helps? Even checking behind on the turn has value in that it gives you a free river card and allows you another chance to catch up. An ace or king may fall on the river, he may actually bet into you and you can raise and win a nice pot, or another blank can fall and you can drop the hand for little loss.
I hope this helps explain the importance of aggression in LHE, and if you take nothing else from this, take these two things:
- Aggression is vital to success in Limit Hold'em. Without playing aggressively, you will almost certainly not be a long term winner.
- Never try to semi-bluff or bluff a calling station. Before you try the plays listed here, make sure your read on the opponent is that he can lay down a non-nut or other mediocre holding. If you feel he can't lay it down, play aggressively, but don't try to overplay him.
Good luck at the tables, all.
4 Comments:
To make sure I'm reading right - are those aggression stats for the aggression factor on the turn? I was curious to compare, so I looked up mine: 1.86.
The link to your auto-rating rules isn't working. even after I took out the extra http the link brought me to Microsoft's home page? Creepy (considering I use Firefox). hehe. I'd be interested to compare your rules to the ones I use (which are actually the ones from the Poker Tracker Guide).
Thank you, Poker Shark. I think I will have to re-evaluate my slumping limit game thanks to your great analysis. Will Wonka tipped me off to this post, so I have to thank him too.
It is awesome when a blogger is willing to share such insight. I can't wait to play limit later.
Dude, another amazing post!
"If you're sure he doesn't have KJ, AJ, or a set or 2 pair (in which case you're screwed), this play works fantastically.
Well, if I could read hands well enough to be sure someone doesn't have those hands, worrying about the intricacies of aggression would be far less important. How can I ever be sure of that????"
yeah i thought that statement was a bit funny. Ok you could say the limper doesn't have one of these hands because you know he would raise preflop and he only limped. You can't be sure one of the other two players that called your preflop raise don't have aces or kings. In fact it's highly likely some of your outs are taken by them.
Also do players actually fold top pair? It sounds like your saying someone is a calling station if they don't fold top pair. I guess if that's the case i'm a calling station - lol. I'd love to play a table of players that all fold any hand top pair or less.
I did find your article very interesting though, especially the aggression stats at the start. One of the best limit articles i've read in awhile tbh.
Post a Comment
<< Home