Quitting as a skill and a hand

I had a short session at Tulalip tonight, which I ended abruptly after losing a big pot where I got my money in very good and decided to stand up rather than continue to play in a moderately good game.

I like to think I have a pretty strong mental game – I’ve lost enough pots to runner-runner outs and I’ve lost to enough improbable Magic topdecks that it’s no longer novel, I’ve never been an angry or particularly emotional person, and I’ve internalized and really believe in the importance of variance in these games for them to be profitable and popular. No matter how much you’ve seen, however, I think everyone is prone to tilt at times. It happens infrequently for me, but it does happen, and one of the ways I’m improving as a player is I’m getting better at recognizing the warning signs and taking action on them.

Quitting a game is an important skill in poker – playing your best game is important and exploiting other players that are off their game is a key skill to develop an edge. Being able to quit a game is a luxury you don’t get in Magic or in tournament poker so it’s a relatively new skill for me which I’m still working on developing.

Here’s the last hand I played in my session:

  • A player limped in from middle position and I made it $25 from the cutoff with xyz9h. The button and small blind fold, the big blind calls, and so does the limper. We’re $500 effective starting this hand.
    I don’t usually play offsuit ace-rags – a while back I cut them out of my preflop range entirely because I was concerned about domination issues and wasn’t confident I could play them well post flop, since then I haven’t figured out how to fit them profitably into my game but it’s an area I am working on. In this case the limper and the players behind me were all weak players that tended to fold too often post-flop and played fairly transparently, so raising in position with a wide range and then continuation betting appropriate flops is going to show a healthy profit.
    I haven’t played with the limper before tonight, who is a white dude in his late twenties/early thirties who has struck me as being a weak player. The big blind just sat down the previous hand and is a weekend regular that I have played with a number of times before. He thinks a bit about what his opponents may have, but only in the most superficial of ways, he’s generally pretty fit or fold, is prone to overvaluing his hand, and doesn’t think about what his opponent’s bets and calls mean.
  • The flop comes As9d2c. The big blind and the limper both check, I bet $40, the big blind raises to $120 (he takes a moment to figure out what his raise sizing is, but he pretty clearly was intending to check raise), the limper folds, and I call after fake agonizing for a moment.
    I accidentally bet a bit too small of this flop – I forgot that I hade raised to $25 preflop and so I thought that the pot was $58, not $73. I prefer betting two thirds pot here – I think most aces are going to be fairly inelastic to bet sizing when they make their calling decisions. I’m betting the flop to get value from other aces – there shouldn’t be many pairs between tens and kings in my opponents ranges, I don’t expect there to be many nines in my opponents ranges (particularly because I have a blocker), and I don’t expect to get value from most other hands. Even though there are only two aces left in the deck, most players’ preflop ranges are heavily biased towards calling with any ace, so they show up much more often in post-flop ranges.
    When the big blind check raises me here he should have a pretty narrow range that I’m solidly ahead of. I don’t expect him to have many bluffs in his range – there aren’t any good draws to semi-bluff with and he’s not the sort to start out a session on a stone-cold bluff. While sets are in his range we can discount them as he’s going to slowplay a lot of the time. He shouldn’t have a set of aces in his range (I expect him to three bet pre), there’s only one combo of nines left, and three combos of deuces, we can probably discount this to one combo total that he would play this way. I’d expect him to play two pair this way and I’d expect him to play ace king and maybe ace queen this way too. That’s around twelve combos that I’m an 85/15 favorite over and a couple of combos that I’m chopping with.
    So we’re happy getting our entire stack in against our opponent and our primary concern should be figuring out how to do that most effectively. I think calling here is the best way to accomplish our goal – the pot is $300 and we have $360 behind, which means pretty much any turn bet from our opponent is going to commit his stack. He might not be willing to get it in on the flop if we shove, letting a turn card peel off is unlikely to kill our action, and he’s unlikely to improve on the turn.
  • The turn is the Th. He bets $150, I pause for a bit, then raise all-in for $200 more. I spend some time refusing to make eye contact while he talks how I couldn’t have ace ten, then talks himself into a call.
    The turn goes pretty much as we planned out – he bets less than half pot, but it’s “a big bet” and I don’t expect him to actually know the pot size, so I don’t think the sizing means anything.
  • The river is the Ks. He flips over AhKc for the rivered two pair.

Hands like this sometimes set off the tilt in me, and I felt it a bit tonight. I was already planning on making this a short session since I have Sunday morning plans and I kind of wanted to go to the gym anyway, so it was fairly easy to stand up. In the few hours since then I’ve been feeling frustrated and I keep thinking back to the hand – quitting was the right play for me and hopefully there’s a bit of catharsis in writing about it too.


Stealing initiative on the flop

I’ve been thinking about continuation bets for a while now. The theory behind continuation bets is that you only make a pair on a flop (with an unpaired hand) 32% of the time, so in a heads up pot about 46% of the time neither player will flop a pair and a bet will frequently take down the hand. Traditionally a continuation bet is made by the preflop raiser, but if we’re out of position and we were not the preflop aggressor, there’s no reason why we can’t make the equivalent of a continuation bet on the right board.

I’ve been thinking about adding this move to my repertoire for a while now and had an excellent opportunity to try it out in a session this week:

  • I’m in a medium-soft seven-handed 3/5 game. The table is pretty weak passive with fit-or-fold tendencies post flop.
  • The player under the gun limps and so does a player in middle position. The guy in the cutoff is about to bet $20 (his standard raise in all positions) but stops, re-evaluates, and decides to bet $25. The button and small blind fold, I call with 5h 5d (we’re deep enough for this to be standard) and the two limpers call.
  • The flop comes 8s 6h 4d
  • I decide to lead out for $60. The two limpers fold quickly, the original raiser looks annoyed and hems and haws a bit, but folds fairly decisively

There are a few things going on here that make it a good spot to donk out and either seize the initiative or just win the pot outright.

  • The original raiser is unlikely to have connected with this flop. The standard bad players at this level are unlikely to raise two limpers with suited connectors (even though at this table it’s pretty profitable to do so) or low pairs, so the only hands he’s going to have that connect with this flop are a suited A8, pocket 8s, and overpairs. I’d further discount the big overpairs (QQ+) because I think he would have thought a bit more preflop and maybe bet a bit more. So we’re ahead of his range.
  • We have some equity if we get called, but not enough to call the original raiser’s continuation bet. We have a pretty easy time bet folding if we get raised, but the original raiser is likely to only do that with overpairs, which is a small part of his range. When we get called we’re probably still ahead, but we also have six outs to catch up. We wouldn’t be able to call any reasonable continuation bet with the 12.8% draw equity we have, but it’s a useful plan B for when we encounter resistance.
  • There is a decent chance one of the two limpers is ahead of us, but they’re unlikely to have connected strongly enough to call our bet. The limpers’ ranges should mostly be suited connected hands and small to medium pocket pairs, which means there are going to be a fair number of single pair hands, but super strong hands are going to be rare. Our line looks pretty strong and the fact that the original raiser has yet to act has to look a bit scary to them. In general when people lead out on flops into the original raiser they tend to have medium strength one pair hands, e.g. top pair medium kicker, so we may see hands as strong as 89 folding here.
  • We maximize our fold equity against the medium strength part of the limpers’ ranges that beats us with this line and we minimize it with a check call line. If we take a check call line we make it really easy for a limper with 67, for example, to call, since they can justify it to themselves that the original raiser is just continuation betting and our hand looks like some draw or something else unthreatening.
  • Taking a check-raise line risks a lot more money and may have even less fold equity than a donk bet does, since the original raiser may become attached to the pot because he’s already put money in. There’s definitely something to be said for making it psychologically/behaviorally easy for our opponents to make the decision that we want them to make.

I like this little weapon, and while I’m not going to make a ton of money with it, I think it’s probably going to turn a tidy profit. There are some things I want to keep in mind when using this in the future, however:

  • It’s important to use this mostly on flops where the original raiser is unlikely to have connected strongly. I don’t want to do this on most Ace high flops, for example, unless I have strong evidence to put my opponent on a pocket pair.
  • I generally don’t want to be out of position but not be the original raiser – in a vacuum limping is not optimal, so I want to make sure I’m only doing it when I have a good reason – so this is more frequently going to be a play I consider when I call a raise out of the blinds (which is also something I hate doing).
  • This play is something that depends heavily on my opponent – I definitely don’t want to run this against completely unthinking players, against stubborn calling stations, against maniacs that like to play back at me, etc.

A tournament hand

Just a quick hand history from a tournament I played earlier this week:

We’re seven handed at the final table and everyone is in the money. The payscale is fairly flat through 3rd place, but there is a decent jump between 3rd and 2nd and a big jump between 2nd and 1st. The blinds are 2,000,/4,000, there is a 400 ante, and I am the chip leader with about 110K (28BB, 12.5 M).

An older player with one of the shorter stacks (20K) limps in. We’ve only been at the same table for an orbit or two, but so far he’s struck me as an intelligent person who is pretty bad at poker. He clearly doesn’t understand push/fold math, is weak/passive, and fairly easy to read. I’d imagine he has some sort of Ace high hand or two broadway cards, but he could have suited connectors or a low to medium pair. I’d heavily discount him having a large pocket pair or Ace King.

It folds around to the small blind, who makes a min raise to 8K. The small blind has the third largest stack, with about 60K. He is also an older gentleman, but has so far struck me as durdly and unthinking. He also doesn’t understand any of the tournament dynamics, and is likely a level zero player, only thinking about his own hand strength. He’s shown some aggression before, and has been open raising 3 to 3.5x the big blind, so his raise here, especially with a limper, is unusual.

I am in the big blind with 2c_e 2s. What’s my play?

If I were heads up against the limper the correct play is to move all in. I have some fold equity, which is significant, but I’m mostly interested in the overlay from the dead money in the pot. If I were called 100% of the time I would be risking 16K to win 28.8K, which means I need about 35.7% equity. I have 40% equity against the range of any pair and any two broadway cards. If we take AA, KK, and QQ out of my opponent’s range I go up to about 42% equity. If my opponent’s range is AJ+, KQ, and JJ-22, which is a pretty tight range here, I have enough equity for a push to be profitable without fold equity.

The presence of the small blind complicates things, however. I’m simply not comfortable going to war with one of the other big stacks with this particular holding – the table as a whole has been pretty passive and I’m confident that I can easily get to top 3 with 20% or more of the chips if I continue to play as I have been (avoiding confrontations with the other big stacks, stealing aggressively, putting pressure on the small stacks). If the SB had just limped in I would likely just check and hope to check it down all the way.

I have the odds to set mine here. There’s 18.8 K in the pot right now, and the SB has 52K back, and I only have to call 4 K, which means I am getting 4.7 to 1 in immediate odds and 17.7 to 1 in implied odds. A clear call if I were heads up against the SB.

But the presence of the limper complicates things yet again. If I call, the limper shoves, and the SB shoves I’ll have to call 52K to win 90.8K, so I would need 36.4% equity. If I’m facing a range of top 20% of hands from the limper and top 10% of hands I only have 27.5% equity. Against two random hands I only have about 31% equity.

If I call, the limper shoves, and the SB only calls I’ll have to call 12K to win 50.8K, so I’ll need about 19% equity, which I probably have. But I won’t always be able to realize all of my equity – sometimes the SB will bet the flop and I’ll either fold the best hand or I’ll fold when I would have made a set on the turn or the river. So while I probably have enough equity to call in this situation I will have trouble realizing this equity.

Finally, if I call and the limper simply calls then I have the correct odds to set mine once again, and I can simply fold on any flop I don’t connect with.

Given the nature of the limper I actually think that the majority of the time I call he’ll call behind or fold. But the frequency with which I get a favorable outcome from calling is so low (call it about 8% of the time, since I flop a set 12% of the time, but sometimes I’ll have to fold preflop to a shove), that it’s not worth the cost, even though the magnitude of the outcome is high (eliminate two players and get up to 190K in chips). If this were a table where I felt I didn’t have a skill edge this would be a good spot to take, but the variance isn’t worth it here, in my opinion.

In the tournament I folded and the small blind shoved on a fairly dry flop. The limper folded and the small blind showed Aces, confirming my belief that his raise sizing was unusual.