Substitute
Jun. 26th, 2006 11:13 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
peterbirksNot really much you can write about when you haven't left the house, haven't read the papers, and haven't done anything interesting. Played one hour on Stars, lost $2.50, making seven red numbers on the trot, session by session. I don't think that's ever happened to me before. In fact, things weren't quite that way, since one of the sessions, on Virgin, was really two separate sessions in a single day — it's just that my spreadsheet counts them as one. And in the second session I regained about $80 of the $210 I had lost earlier.
And then I played an hour on Party and, stone me, won at three tables out of three. So, that was the second step of the one step forward, one step forward, mode (although the first part of Sunday was four steps back, so, still some way to go). Sisyphean, indeed.
I compiled a spreadsheet of my performance with Ax, suited and unsuited, in the small blind. Ignoring AK, which is a special case and needs special treatment, and recalibrating all the sites to $2-$4 (I haven't included the $5-$10 stats because that, too, is a different game), I came to the following tentative conclusions.
1) If you can get in for a limp, any Ax is worth playing providing your post-flop play is competent.
2) Only AKo is worth calling a raise for. If it's a steal possibility and you have AJo or AQo, then you have to three-bet. AKo can be three-bet as well (whether it's a steal or not), but not every time.
3) AKs, AQs and AJs are worth calling a raise for. ATs, A9s and A8s are possible three-bets to a potential steal, but you are on risky territory here, because ATs isn't that hot against AJoff heads-up. Unless the guy is a manic stealer on virtually anything, I'd be hesitant about playing ATs or A9s in this situation. I think that part of my leak was calling raises with these "half-way" Ace hands when the actual odds weren't there. You probably need four or more opponents and little chance of a three-bet to call a raise in the SB with this kind of hand.
Stats were:
Hand Times Won Net VPIP Gross if passed Difference
ATs 33 31.25% ($35.66) 70.83% ($33.00) ($2.66)
A9s 35 30.36% +$63.80 80.36% ($35.00) +$98.80
A8s 30 15.99% ($60.30) 79.01% ($30.00) ($30.30)
There were more stats available on the offsuit hands. Over a sample of 1126 hands, my gross was a loss of $661 (recalibrated to $2-$4), which was $426 better than it would have been if I had passed every hand. But it was still a loss of 0.15 Big Bets a hand, as opposed to a target (for all SB hands) of minus 0.11. Once again, A8off seems to be the danger hand. Perhaps I'm not getting away from it when I should (e.g., flops of 8xx and Axx where I meet resistance?). I'm certainly not calling raises with this hand, so I suspect that I'm getting caught by hands like AJ AT and A9 which have limped first in, or some such nonsense.
+++++++
One of the powers of Pokertracker is that it gives you a good clue of when to slow-play a big pair from early position (or even later position).
You may remember the absinthetics post recently, asking what could be done about a type of player who raises more and more liberally in later and later positions, and then runs a kind of aggressive-passive algorithm if he is called.
Well, part of the solution is for there to be other players around who know what to do. Basically, if you are in the big blind and this guy is in the cut-off, you have problems unless there is a quality player in early and/or a quality player in late.
If you are the quality player in early position, you can limp far more often with powerhouses. Many is the time you will be raised. You do not three-bet pre-flop. You go into check-call mode until the river, at which point you value check-raise unless there is an overcard. With something like QQ and a board of Kxxxx, (no possible flush or likely straight) you might even value check-raise here.
However, if you have a power pair in early position and the players behind you are more standard $2-$4 fare, either a collection of weak-tights 9%/%5%s or loose passive 38%/4%s, and there is a very loose player in either the small blind or big blind, a limp with a big pair is a bad mistake. First, you are likely to get either very few limpers behind you (in a tight game) and then you allow the blind in for free, whereas he would defend his blind with any two cards, or you are likely to get a school of callers behind you. This is slightly less negative EV, because sometimes your KK will hold up or you will flop a set. But your out-of-position situation and a mass of minefields behind you mean that, in effect, you have turned your KK into something approximating a middle pair.
So, that's the limp-call and the "don't limp" situations. When should you limp-reraise? Well, preferably with a bunch of maniacs who aren't paying attention, or when you have a maniacal image. Because any other time, you might as well hold up a "I have Aces or Kings" sign. The best position to do it is with a very aggressive player on your immediate left and a bunch of weak-tights behind him (plus at least one tight player in the blinds). In this situation a limp reraise might be best if you get a caller in the blinds. This gets in two (or three or four, if our aggressive player caps it, as he well might in an attempt to eliminate the caller in the blind)) extra bets before the flop. If the blind gives up to the cap bet from the aggressive player, you can bet out on any flop (including an Ace). If it comes Axx, life is tougher. These aggressive players often like to raise on Ax, whereas blind defenders also like Ax. You may well be in difficulties here and folding might be the best option. However, if there is no caller in the blinds and you are heads up pre-flop, I prefer a call. Check-call the flop and, provided no Ace appears, check-raise the turn.
Now, back to our liberal raiser. What do you do if you are a quality player in late position? Well, obviously, you three-bet him with far thinner values and, fun fun fun, flat-call him with potential multi-wayers when there are looseish players behind you or in the blinds. I'm not a great fan of cold-calling raises in late position. Usually it's a case of reraise or fold. But here you are fighting a specific kind of opponent, and you need to adopt unusual tactics. This is a variant of the Caro "call rather than raise" approach. Since a large number of daytime $2-$4 players (incorrectly) flat-call raises with hands like JJ and QQ, and then put in a raise on a rag flop, but flat-call if an overcard is on the board, your liberal raising opponent will not be quite sure where he is with you. If you flat-call the raise with, say, T9s or 98s, you aim to get away from the hand if it becomes a multi-wayer and you fail to seriously hit, but if it is heads-up, you are willing to call the man down with any pair, provided there is no more than one overcard (in this case, something like KK953 counts as one overcard). Obviously, if you hit two pair or a straight, then you will raise him on the river. Remember, you know this man's game plan. He is going to continue betting at you. That is your strength and his weakness.
And then I played an hour on Party and, stone me, won at three tables out of three. So, that was the second step of the one step forward, one step forward, mode (although the first part of Sunday was four steps back, so, still some way to go). Sisyphean, indeed.
I compiled a spreadsheet of my performance with Ax, suited and unsuited, in the small blind. Ignoring AK, which is a special case and needs special treatment, and recalibrating all the sites to $2-$4 (I haven't included the $5-$10 stats because that, too, is a different game), I came to the following tentative conclusions.
1) If you can get in for a limp, any Ax is worth playing providing your post-flop play is competent.
2) Only AKo is worth calling a raise for. If it's a steal possibility and you have AJo or AQo, then you have to three-bet. AKo can be three-bet as well (whether it's a steal or not), but not every time.
3) AKs, AQs and AJs are worth calling a raise for. ATs, A9s and A8s are possible three-bets to a potential steal, but you are on risky territory here, because ATs isn't that hot against AJoff heads-up. Unless the guy is a manic stealer on virtually anything, I'd be hesitant about playing ATs or A9s in this situation. I think that part of my leak was calling raises with these "half-way" Ace hands when the actual odds weren't there. You probably need four or more opponents and little chance of a three-bet to call a raise in the SB with this kind of hand.
Stats were:
Hand Times Won Net VPIP Gross if passed Difference
ATs 33 31.25% ($35.66) 70.83% ($33.00) ($2.66)
A9s 35 30.36% +$63.80 80.36% ($35.00) +$98.80
A8s 30 15.99% ($60.30) 79.01% ($30.00) ($30.30)
There were more stats available on the offsuit hands. Over a sample of 1126 hands, my gross was a loss of $661 (recalibrated to $2-$4), which was $426 better than it would have been if I had passed every hand. But it was still a loss of 0.15 Big Bets a hand, as opposed to a target (for all SB hands) of minus 0.11. Once again, A8off seems to be the danger hand. Perhaps I'm not getting away from it when I should (e.g., flops of 8xx and Axx where I meet resistance?). I'm certainly not calling raises with this hand, so I suspect that I'm getting caught by hands like AJ AT and A9 which have limped first in, or some such nonsense.
+++++++
One of the powers of Pokertracker is that it gives you a good clue of when to slow-play a big pair from early position (or even later position).
You may remember the absinthetics post recently, asking what could be done about a type of player who raises more and more liberally in later and later positions, and then runs a kind of aggressive-passive algorithm if he is called.
Well, part of the solution is for there to be other players around who know what to do. Basically, if you are in the big blind and this guy is in the cut-off, you have problems unless there is a quality player in early and/or a quality player in late.
If you are the quality player in early position, you can limp far more often with powerhouses. Many is the time you will be raised. You do not three-bet pre-flop. You go into check-call mode until the river, at which point you value check-raise unless there is an overcard. With something like QQ and a board of Kxxxx, (no possible flush or likely straight) you might even value check-raise here.
However, if you have a power pair in early position and the players behind you are more standard $2-$4 fare, either a collection of weak-tights 9%/%5%s or loose passive 38%/4%s, and there is a very loose player in either the small blind or big blind, a limp with a big pair is a bad mistake. First, you are likely to get either very few limpers behind you (in a tight game) and then you allow the blind in for free, whereas he would defend his blind with any two cards, or you are likely to get a school of callers behind you. This is slightly less negative EV, because sometimes your KK will hold up or you will flop a set. But your out-of-position situation and a mass of minefields behind you mean that, in effect, you have turned your KK into something approximating a middle pair.
So, that's the limp-call and the "don't limp" situations. When should you limp-reraise? Well, preferably with a bunch of maniacs who aren't paying attention, or when you have a maniacal image. Because any other time, you might as well hold up a "I have Aces or Kings" sign. The best position to do it is with a very aggressive player on your immediate left and a bunch of weak-tights behind him (plus at least one tight player in the blinds). In this situation a limp reraise might be best if you get a caller in the blinds. This gets in two (or three or four, if our aggressive player caps it, as he well might in an attempt to eliminate the caller in the blind)) extra bets before the flop. If the blind gives up to the cap bet from the aggressive player, you can bet out on any flop (including an Ace). If it comes Axx, life is tougher. These aggressive players often like to raise on Ax, whereas blind defenders also like Ax. You may well be in difficulties here and folding might be the best option. However, if there is no caller in the blinds and you are heads up pre-flop, I prefer a call. Check-call the flop and, provided no Ace appears, check-raise the turn.
Now, back to our liberal raiser. What do you do if you are a quality player in late position? Well, obviously, you three-bet him with far thinner values and, fun fun fun, flat-call him with potential multi-wayers when there are looseish players behind you or in the blinds. I'm not a great fan of cold-calling raises in late position. Usually it's a case of reraise or fold. But here you are fighting a specific kind of opponent, and you need to adopt unusual tactics. This is a variant of the Caro "call rather than raise" approach. Since a large number of daytime $2-$4 players (incorrectly) flat-call raises with hands like JJ and QQ, and then put in a raise on a rag flop, but flat-call if an overcard is on the board, your liberal raising opponent will not be quite sure where he is with you. If you flat-call the raise with, say, T9s or 98s, you aim to get away from the hand if it becomes a multi-wayer and you fail to seriously hit, but if it is heads-up, you are willing to call the man down with any pair, provided there is no more than one overcard (in this case, something like KK953 counts as one overcard). Obviously, if you hit two pair or a straight, then you will raise him on the river. Remember, you know this man's game plan. He is going to continue betting at you. That is your strength and his weakness.
