Wednesday, May 09, 2007

Tournament Strategy

I've received some good emails lately and today I thought I would share something I read and give my thoughts, and then solicit yours as well. So another poker blogger sent me the following hand history a couple of weeks back. The bubble in the weekly 400k guaranteed on full tilt just burst a few hands ago, so you (Hero) are already in to the minimum payouts. Your stack is 22,440, giving you an M of around 8 with blinds of 600-1200 and an ante of 150. You are solidly in the middle of the pack and are in good shape to be patient but aggressive:

600/1200 Ante 150
Seat 1: Hero (22,440)
Hero posts the big blind of 1,200
*** HOLE CARDS ***
Dealt to Hero [3d Tc]
1creeper calls 1,200
Hero checks
*** FLOP *** [Ts 5c 6d]
Hero checks
1creeper bets 3,600
Hero calls 3,600
*** TURN *** [Ts 5c 6d] [4s]
Hero checks
1creeper bets 7,200
Hero raises to 17,490, and is all in
1creeper calls 10,290
Hero shows [3d Tc]
1creeper shows [Jd Td]
*** RIVER *** [Ts 5c 6d 4s] [6c]
Hero shows two pair, Tens and Sixes
1creeper shows two pair, Tens and Sixes
1creeper wins the pot (46,530) with two pair, Tens and Sixes
Hero stands up

This blogger then followed up with the following comment, which is what I really found interesting. FWIW I had commented about this hand that I thought it was unnecessarily slow-played and then hung on to despite a terrible kicker:

"Thanks for the comment. I don't like how I played the hand either.
The reason I found it interesting is because it's one of those strange tournament situations where I try to "get lucky" or "buy the pot" while "capitalizing on small edges." I mean, this is what I don't understand about tourneys. If you have to get lucky, if you have to build a stack, if you think you have around a 50-50 shot of winning the hand after combining odds, fold equity, etc...then why shouldn't you go for it?
These questionable hands come up rarely in tournaments, and "waiting for a better spot" is a recipe for stack shrinkage while blinds continue to rise.
I don't know. Obviously I donked away my entire stack when it wasn't necessary, and it wouldn't have been hard to fold.
You can't win an MTT by folding though. So what am I missing?
If I had to do it over again, I think I would have check-raised the flop rather than check-calling the flop and check-raising all-in on the turn. If I do that, I increase the chances that my opponent would fold to my turn bet.
Anyway, I don't understand tourneys. I don't know why people limp from MP with JTo. I don't know why they call an all-in with it when they could easily be dominated on the turn."

OK. There is a lot to cover in there. Let me start by addressing the first overall point I think you're making, about why not take a race at a 50-50 chance to double up or go home. This is interesting because most of you who have played mtt's for a while, I would bet you that you've all struggled with this at some point. After a while, especially after a nice win or nice cash or two, you find yourself just chasing every race you can, reasoning that you'd rather just double up early or go home early. That is an acceptable position, as long as you really don't mind the idea of going home early. That's because by taking up every race situation you can find, you will be significantly increasing the chances of going home. Period. Mathematics doesn't lie, and this is math that even I remember the details of. Constantly putting yourself in race situations does not pay over the long run, if your goal is to survive to the end in tournaments, as is my goal every single time I play. My goal is never to "double up early or donk out early". It's to survive, play smart, aggressive poker and just keep going, keep alive.

I assume this all is very obvious to most of you, but just to put this discussion in perspective, let's say over the course of each of the first two hours of a large mtt, you call allin, or raise allin against a guy you are fairly sure is going to call you and he does, two times, and both times in each hour, for a total of four races, are race situations. Either you have AK against his 77, or you've got the 77 and he's got the AK or AQ. Assuming an exactly 50% chance of winning each of these, there is a 50% chance of you surviving the first race. Now, while the chance of you also surviving the second race (or any individual 50-50 race, for that matter) is still 50% of course, the odds of you winning that race times the odds that you are still alive having won the first race already are 50% times 50% or 25%. So you have a 1-in-4 chance of winning two 50-50 races. 75% of the time, when you take the first race and stack off on that second race by calling allin with your AK preflop when you know the reraiser has a pair of some kind, you're done. In a tournament which by definition rewards only longevity, you have just automatically put yourself in a situation where you're out in the first hour alone 75% of the time. Hmmm.

Then in the second hour, the odds of you being alive at all after the two first-hour races still are only 25% as we just covered, and on top of that you take on a third 50-50 race early in the second hour. Now your odds of surviving past the beginning of the second hour are down to 12.5%. 1 time in 8. 7/8 of the time you follow this pattern, of just two races an hour, 7/8 of the time you are automatically out of the tournament in the first 80 or 90 minutes. Then if you go for one more 50-50 race at the end of the second hour to try to "chip up or go home before the break" (we've all been there), now your chances of surviving all four of these 50-50 races have dropped to 6.7%. One in sixteen. Basically speaking, that means you ain't making it to the third hour. No matter what else you do outside of these four races, no matter how many other chips you win with good plays. You willingly take on four allin races during 200 hands over 2 hours of play in an online mtt, and you've cut your chance of surviving to the third hour down to 6.6% just from those races.

And keep in mind, these are not all totally donkish-looking plays you're making, mind you. This is like when you are dealt AK in the big blind. Late middle position open-raises 3x, and the button reraises his bet 3x. That guy is almost certainly on some kind of a decent pair. Maybe AK, but since you have an Ace and a King, it's much more likely that he's got a decently high pocket pair, like QQ, JJ or TT. Anyways, you re-reraise your last 400 additional chips on top of his 900 chip bet, knowing he has to call with the odds to take on a race against your AK. Is that the worst move ever? No, not by a long shot. But, it is roughly a 50-50 chance of allin winning or allin losing, since you know he's not folding for another 400 chips. In a tournament, those who get rewarded are those who last, who protect their stack, and grow it while protecting it effectively at the same time. That means taking some chances to chip up, but not putting yourself in basically total luck's jeopardy of being busted completely. 6.6% chance of survival until the third hour is not protecting your stack.

What you will find if you continue to go after all-or-nothing races early on in mtts is that races aren't that hard to find if you're really looking for 'em. Trust me. I've been there. For a while last year I let success get to my head and I chased every chance I could to double up or just start up another mtt five minutes later. And you know what happened, eventually? At first it was fun, your memory likes to remember the big double up suckouts and huge early stacks more than all the donkouts, but it's enjoyable at first. Then after a month or two of this strategy, you will be like "wow, I haven't been to a final table in ages!" It's happened to many of us. You can't go out looking to get allin preflop early in an mtt, unless you don't mind significantly increasing the chances of an early exit. It's not just a 50% chance. That's just one race, and it's not even counting all the other pitfalls -- the set over sets, the rivered four-flush, that you have to endure in mtts. You take on two races, and your chances of survival are 25% or less. Repeated 50% chances to lose everything is no way to protect your stack, your roll or any of your assets.

Now to be sure, there are times in every tournament when taking an allin race might be a great idea for you. Mostly in my experience it is at the extremes of stack sizes. If you have a very large stack, you can certainly consider calling allin with 8s when a late position shorty open-raises allin with what is probably any Ace in his position. Why not, you can afford to lose if along with that comes a 50% chance of adding more to your growing stack. You are favored ever so slightly, after all. I do that all the time, and it's an essential part of being a bully, and continuing to be a bully, in later-stage mtt play.

Similarly, if you yourself are very short (or in some cases even if just the other player is very, very short), then of course a 50-50 chance to either double up or go home, the latter being what you are very close to doing anyways, can become an attractive option in a big hurry. You surely cannot wait for pocket Aces, pocket Kings or AK when you're down to 5000 chips at a 5-handed table with blinds of 600-1200 and a 200 ante. In that spot, with an M that low, a double up by calling a preflop MP raise allin with your AK is probably a great play. However, with an M of 8 and in the middle of the pack having just made the cash (probably around $250 or so) in the 400k, I don't think you are anywhere near race-calling time yet.

Now moving on to your next point, about you trying to "get lucky" or "buy the pot" while "capitalizing on small edges", let me say for the record that I am totally with you on that. Totally. That is exactly what you have to do in mtts, especially as play progresses into the late stages where the Ms are low and the blinds are sky high. However, I would suggest that a desire to try to "buy the pot" and "push the small edges" does not lend itself to (1) slow-playing top pair on the flop, nor (2) continuing to push in the rest of the money into the pot with an easily beatable hand.

First, about number (1) above, when you flopped top pair, why not bet it out there? Yes your kicker is a 3, but why not put in a nice bet there? If he calls you, you have a tough decision, and if he raises you then you can happily fold, having lost the minimum given your situation I would say, because you know your kicker won't even play. I think pushing your small edges and "buying the pot" dictates betting any top pair on the flop in that scenario, so I support that. Check-calling the flop made it impossible for you to "buy the pot" since you never bet at it, and with a hand as weak as top pair Tens and an unplayable 3 kicker, you instead allowed your opponent to stick around and/or get more money from you if he is actually ahead. Here if you bet the flop, you make your attempt to buy the pot but still leave yourself an out to escape from the hand if he calls and certainly if he raises you, with you getting away from a potential elimination hand for only a minimal loss. The way you played it, you basically decided on the flop that your top pair Tens and 3 kicker was definitely best, then allowed him to do whatever he wanted and you were not going to change your mind about your hand being ahead, and in doing so you ended up playing it kinda passively until your opponent was heavily committed before your big checkraise on the turn. That kind of slow-check-and-call-and-then-checkraise-the-turn move is like the total opposite of buying the pot the way I see it. So I do support your strategy in late-stage mtt play of "buying pots" (you have to all the time if you expect to make it real deep in a big mtt), but I generally just try to take them as quickly as I can when that happens, when I figure my chance of winning or of finding out if I'm behind for cheap is the greatest.

About point (2) above, I think once he bet the turn after you called on the flop, you've got to start thinking he might have a hand. Is he likely to bet out again with middle pair after you called his flop bet? It's certainly possible, no doubt, but I would estimate it to be more likely that he has some kind of a real hand, and against almost any "real hands", you are way, way behind at this point. I'm not saying I would just automatically check-fold on the turn, but that doesn't mean I would willingly get it allin with top pair 3 kicker against a guy who's already shown a willingness to voluntarily commit thousands of chips to this pot, even a second time after you called him the first time. That's pretty strong right there, and your hand fares very poorly against any kind of very strong holding from your opponent here. See, this is why I would have bet that hand out on the flop. Because by the time the pot is big on the turn, you will know whether or not to commit any more chips at all to this pot, if you're even still in the pot at all. By waiting until after his bet on the turn, you are now faced with calling off most of your stack, or even reraising allin like you did, without having had a single chance to try to figure out how good his hand really is, which is very important since your hand is among the worst possible hands that can be considered good at this stage.

When I flop top pair 3 kicker, I'm betting out on that flop probably 95% of the time. I'm betting on the flop a high percentage of the time anyways, so this helps camouflage things a bit, but you do not want to see a free turn card with top pair shizza kicka. I want to bet that flop, and find out early while the pot is small whether my top pair shizza is ahead or behind. And my tournament game is easily at the point where I lay down that top pair shizza kicka readily to any showing of strength. Lots of people play Tens, and I've played enough heads-up over the past several months to know that two people are dealt "dominating" hands together more often than you might think. So I think I would have lost the minimum there, while at the same time doing my best attempt at a quick "buy the pot" move right there on the flop. Check-calling the flop and checkraising what is then a bigger, scarier bet on the turn is not where I suggest you want to be with top pair no kicker.

Getting to your last points, you said: "Anyway, I don't understand tourneys. I don't know why people limp from MP with JTo. I don't know why they call an all-in with it when they could easily be dominated on the turn."

Sorry guy, but I can't help you there. JTo is not a hand I am apt to limp in with from middle position under any circumstances in a tournament. Calling allin with JTo if that is just top pair on the turn is probably not a profitable move over the long run, but I will say this. If your opponent in that hand was any kind of a player, he would have been willing to get away from top pair Jack kicker under the right circumstances. But you gave him no reason whatsoever to get away from the hand! Think about it, by check-calling the flop, you gave the impression of something weakish -- callable, but barely. Something like middle pair decent kicker, top pair bad kicker, maybe a poor draw, something like that. Then when you checked again on the turn, you further solidified that read in his head. Level 3 thinking is all about "what does he think I have?" In this case, using level 3 you should understand that by your play you've been telling him you've got something weak, and now you're suddenly raising allin on the turn after telling him you're weak twice in a row? I probably would have called that bet too, or I certainly would have given it serious thought. I mean, I concede that sometimes you play a flopped set that way, but I would probably be much more apt to put you on exactly what you had here -- top pair nothing kicker -- or maybe two pairs or a straight on that turn card. His call certainly was not the best move I've ever seen by a long shot, but I think it doesn't make sense for you to play it weak like you did all through the hand, then suddenly spring the allin raise on the turn, and then wonder why he called with just top pair decent kicker.

Let me pose the following scenarios, any one of which I think is a better way to play this hand, and hands like this in general in late-stage tournaments, and one of these is more like the way I would aim to play a hand like this in that spot:

1. You hold T3o and the flop is T65. You bet the flop, and he raises.
Outcome: If you're a man, you fold. You lose just one small bet on the flop.

2. You hold T3o and the flop is T65. You bet the flop, and he smooth calls. On the turn you check, and he bets.
Outcome: If you're a man, you fold. You lose just one small bet on the flop.

3. You hold T3o and the flop is T65. You check the flop, planning to checkraise with top pair. He bets, and you raise. He reraises allin.
Outcome: If you're a man, you fold. You lose one raised bet on the flop.

4. You hold T3o and the flop is T65. You bet the flop, and he smooth calls. On the turn, you bet again to represent a strong hand to try to get him to lay down his top pair, and if he raises, you reraise allin. Then if he calls you with just top pair Jack kicker, he is arguably a dolt. In that situation you bet the flop, and then you bet and reraised on the turn, so you've screamed to him that you're strong, even in the face of him having called your flop bet and raised you on the turn. He has to put you on a very strong hand at that point, and in that situation, if he is a man, he is folding. I can't say that he will fold every time, but he should fold because you've given him every reason to believe his top pair Jack kicker has got to be behind. Instead, in your situation you told him you were weak on the flop, and then you checked to show weakness again on the turn. The checkraise does work some times, but in general I think if the plan was the checkraise on the turn, then doing something to show some strength before the allin move will only help your cause.

And yeah, people limping JTo from MP is sick and wrong.

Hope this was helpful or at least interesting for some of you out there. Don't forget the Mookie is tonight at 10pm ET, and I hope we can set a new record there after setting all-time attendance records at the last MATH as well as Don's last Big Game as part of the BBT. I will 100% definitely be following the action on Lost during Mookie time, but I suppose it is possible that I will be in there as well. For the past few weeks, I've just donated as I've sat the computer off on the side, and only looked up to play maybe 5 or 6 hands in the entire first hour while Lost is on. But I love the Mookie and I like to donate to the BBT cause, so I've been in there doing my part, and one of these days I will hopefully find a way to survive until the second hour and actually get to start playing.

And speaking of Lost, tonight's show as well as the last couple, are all clearly going to be amazing. I guess tonight is when we find out what Jack and Juliette's little secret is -- which I am telling you right now is about another member of the Losties and not at all something about the island or the Others, you heard it here first -- as well as Ben allegedly telling Locke about "the secrets of the island" according to my DVR summary. That sounds like the recipe for a balls-kicker of a show if you ask me, so no way I am missing that shiat tonight. So get your multi-tabling skills working early on Wednesday evening so you can watch the best episode of the year in the best show on television (24 sucks nowadays, that formula got old in a hurry huh?) while you call preflop allin raises with QJo on your way to final tabling the Mookie. See you then!

Labels: , , ,


Blogger Mike Maloney said...

Good stuff Hoy, very helpful. Interesting to think about the idea of pushing all-in early on in a tournament considering what happened at the first hand of the MATH this week. Speaking of which, I've been wanting to talk to you more about that hand, since you thought it was a crappy play, but I don't have your e-mail or girly chat screen name or any other way of talking to you outside of leaving you a comment. But I would like to disucss the hand further, and see where I made a mistake in my thought process.

10:49 PM  
Blogger oossuuu754 said...

Good Read, keep up the great posts.

10:54 PM  
Blogger Hammer Player a.k.a Hoyazo said...

To be honest, Mike, I've played so many mtts at this point that I've learned to keep my memory so short that I don't even recall the hand you are referring to. But I'd love to chat about it. I think I am "sandlerm" on the yahoo girly chat, come hit me up this evening and I should be around. Except during Lost.

10:55 PM  
Blogger bayne_s said...


In item 4 of how to play the T3 hand you have left out make sure other player is not "Waffles". There is no chance waffles is laying down top pair J kicker or top pair T kicker once his powerful JTo has connectted.

He is also player most likely limping JTo in MP.

12:32 AM  
Blogger smokkee said...

T-3o is a loser.

i've been knocked out of too many tournaments trying to defend my blind with marginal hands. this is a good example of that.

if he had just open pushed the turn, the other clown may have folded. but, maybe not. not to pick on wawfuls ;). but as Bayne said, wawfuls is definitely not folding TP there.

with that many chips, you're better off looking for a another spot to get all ur chips in there.

1:25 AM  
Blogger Eric a.k.a. Bone Daddy said...

Great questions and I agree with your table of how to play out each hand.

My thoughts are less about the specifics, I think people in MTTs panic too much with Ms between 5 and 10. Everyone has a shitty M at this point, sans a few fatties, and all it takes is one double up from them to get healthy and put you in position to pressure the shorties.

So blind fights at this point are stupid. Continued aggression, as you sugggest, is the best way in my opinion, by raising pots that will help maintain a livable M until you hit a hand without risk of losing your entire stack with so so holdings.

Just curious, what do the women in the crowd do when they want to fold?

1:26 AM  
Blogger TechGuyTom said...

Lets start from the bottom and work up.

Limping with JTo in middle position is not wrong, it's a style choice. The hand history doesn't give stack sizes, but given a decent stack I would never fault that play preflop. If you can get away from a hand and you have a decent read on your opponents, any two cards are good.

The scenarios are good, nothing to add.

I agree with your advice about betting out on that flop. Check calling is the worst decision to be made in that hand IMO.

I am going to disagree with what I think you were trying to get across with the percentages discussion. Granted I have the attention span of a gnat so I tend to gloss over your novella length posts :) This may lead to a misinterpretation on my part.

I believe that you were trying to discourage the practice of taking coin flips based on the fact that previous coin flips were taken. This approach will lead to "weak" thinking. My approach is to avoid coin flips if possible but if the situation warrants it, one should never disregard the action simply because they won the last two or three coin flips. Every situation is a reset. The results of a previous hand can in no way influence the results of this hand. In essence every situation starts with a new 50/50. Not a 50x50=25.

1:47 AM  
Blogger Hammer Player a.k.a Hoyazo said...

L.A.G., I have clearly done a bad job of getting across my point about the percentages of winning all those coin flips. Surely the chance of winning any given 50-50 coinflip is always going to be 50-50. Nothing can change that. However, no one also can change the fact that the chance of winning four out of four 50-50 coinflips is 1 in 16 or 6.666%. That's just basic math.

So it's not that the chance of winning any given coinflip is changing depending on what has occurred in past coinflips. What I'm trying to say is that the chance of winning 50-50 on all four out of four coinflips during the first 2 hours of play is, in fact, exceedingly low, and thus what follows is that looking to do a couple of coinflips an hour in a large mtt makes it practically almost impossible (mathematically speaking here, not opinion) to survive to the cash.

Clearly you are 100% correct though, the results of previous coinflips cannot possibly enter the decision of whether to take on a new coinflip situation or not. That's why the calculation is always a 50% chance for each coinflip. But four of them still equals 50% x 50% x 50% x 50% or 6.666%. This is why going for the early coinflips in a large mtt is not the right long-term mtt strategy IMO.

Thanks for the comment and the clarification.

1:55 AM  
Blogger slb159 said...

thx for stopping by and commenting...well said.

as a sort of "beginner" at the tables, I post what I see, and most of the time, what it results in.

I trust the majority of my readers are beginners too and just giving them my observations.

again, well said though, I should stay away from "Don'ts", "Never's" et cetera, since no move in poker is ALWAYS the correct one.

Take care

2:06 AM  
Blogger WillWonka said...

Nice post, as usual. I had a similar type hand where I tried to "buy a pot" with a marginal hand. I wrote a little about it today and would love your thoughts..

Basically a MP limper going heads up with me (I'm BB with K5) on a 757 board.. I check raise all in..

What can/do you normally put people on with MP limping.. obviously the range is very wide.

Keep up the good stuff.

4:04 AM  
Blogger Gnome said...

Thanks for the insight and the post. A couple of thoughts:
_ On the argument about the diminishing returns of taking coinflips, I'm not sure the math quite works out so neatly. The reason is that your risk or ruin is not equal on each successive coin flip. For example, if you win the first coinflip and all other stacks are equal, you have 3,000 chips and the rest of the field still has 1,500. Then if you lose the next coinflip, you're stack only diminishes to 1,500 again -- not 0. I know what you're saying though, and I'm (finally) coming around to the idea that coinflips are evil.
_ As you say, betting out the flop would have worked best. I was feeling trappy with a mediocre hand, and then I shot myself in the head when I picked up that damn OESD on the turn.
Still, I find it difficult to fold made hands with drawing possibilities against some of these donks who could have anything. I guess I need to respect my opposition more, but it's hard to respect them when they limp in with crappy unsuited connectors.

8:19 AM  
Blogger Crispin said...

Good post. I only recently started reading your blog.

Coinflips can be good in certain situations, ie. you're short stacked or you're headup vs a better player or you're against a short stack that won't hurt you if you are in a coinflip... But if you have a decent stack, why risk a coinflip that could cripple you.

Play tight, aggressively and pick your spots to commit your chips, but also survive to make those final tables.

9:46 AM  
Blogger TraumaPoker said...

This situation reminds me of what you hear over and over again about not going broke in an unraised pot. He can limp a wide range there that has us beat, just because he made the money in the 400K doesn't mean he is an allstar, as we see by the limped JTo.

8:07 PM  
Blogger TripJax said...

They've done studies on all-in coinflips..."50% of the time, they work everytime."

(I tink I quote Anchorman on a daily basis now...)

But seriously, I had actually come to convey the point Gnome was working at in his 1st paragraph, but I think he did it well.

And this is absolutely not to disagree with you, but just to state that I think I take a different approach to coin flips. I want to avoid them early when I can, but I'm not afraid to take them if the situation warrants it. I don't ever think, "wow I JUST took a coin flip, so I better not take this one now even though I feel it is correct under the proposed situation."

And I don't think you are saying that either...just conveying my thoughts and such.

To me, people who fear being all in - whether it is on the 1st hand, on the bubble, or on the last hand of a tourney - will ultimately fail MTT poker. Not to say that they can't make money and play for a long, long time, but that they will not be playing to their max potential. Anyone who fears a particular situation is missing out on so much.

Take a batter in a baseball game vs. an erradic pitcher. If he is scared that he will get hit on any pitch, then he probably won't do well in most at bats with that pitcher. Controlling his fear and not letting it change the way he attempts to hit that ball is so important.

Same thing in poker. If you are scared at any point in poker, you are costing yourself in the long run. Make the correct decision for the current situation at hand - without fear of the outcome - and I believe you will succeed in the long run. Though it doesn't happen all the time, I want to be able to say I made the right decision in any situation that was presented to me.

So anyway, I think this was a great post and I totally got off the subject and have been rambling, but thought I would comment.

I agree with what you've written and wanted to add my 2 cents...

4:10 AM  

Post a Comment

<< Home