Hypotheses not relating to angels on heads of pins

[peterbirks]

I played badly late yesterday evening. This was a standard fault -- assuming that other players are cleverer than they are. It's a habit you can get into if you play too often, because you feel instinctively that opponents should by now be "learning", so you try to zag in anticipation. But this isn't high-level poker. You have lots of opponents, not one table's worth. Any change in their skill level will be a slow alteration, not a sudden shitft.

One example of poor play was when, in the small blind, I raised three limpers (50c-$1) to $5.50 with KK. One very loose passive player (53%/4%, something like that) called from MP1 and the two guys behind him folded. He was also quite short-stacked, down to $20, so that left him with $14.

Flop came A73, two of a suit, something like that. Anyhow, I contrived to double him through with his A5o when I know that these guys will check it down if they have a lower pair. But, well, I thought to myself, "perhaps he's trying a desperation steal here". I'm check-calling with AK here, so I chose to check-call his $5 flop bet with KK, on the grounds that this would lead a bad Ace to check it down on turn and river. He didn't.

In a couple of other cases I called guys down because they were in an ideal situation to "make a play". Needless to say, they weren't making a play. Their hands were perfectly represented by the size of their bets (check = poor; small-to-medium = not sure; big = good). Bringing $200 buy-in thinking to $100 buy-in tables late in the evening is moronic.


Here's a situation based on a hand that cropped up last night.


BB: Tight-not-very-ag (8%/4%) player whom I've played a fair bit (abut 500 hands together): $104

Button: Me, $78 Hand: 3d 3h

Blinds 50c - $1

All passed round to me. I raise to $4.50

BB reraises to $11.


Hero: ?


If you choose to fold, end of story.

If you reraise another $11, he calls and will check any flop. Given that you now have $56 behind and the pot is $44, how do you act on the flops below?

If you call, opponent will bet $14 on each of these three flops. How do you react in each case?


A 7 2 two of suit

T 8 6 two of a suit

Q Q 5 rainbow.


A lot of potential scenarios there, I know, so I expect a deafening silence. Posts are only generated as a rule when the respondent has the benefit of knowing what happened.

However, these are far more important than the previous AK "angels on the head of a pin" scenario, because situations such as the above happen very often, and they happen far more often in $200 buy-ins than $100 buy-ins and far more often short-handed than in ring games. Coming to the right EV decisions in these situations matters many times more than in the previously mentioned hand. And yet these hands generate little interest.


What does this imply? Probably that most poker players are looking in the wrong places to improve their game.

Which is fine by me.

++++++++++++++

MR PICKLE UPDATE

Ohh, it never rains but it pours. Not only does a minority shareholder threaten to kybosh the recommended Northern Rock deal, but it looks as if the NTL e-mail system has completely collapsed this morning. All e-mails to virgin.net people this morning are bouncing back and NTL broadband customers have been reporting zero access. Oh dear, oh dear.

Comments

Half an answer

[(Anonymous)]

I'm fairly sure that in the scenario you give I fold straight away. His figures suggest a hand he is at least prepared to see a flop with, unless you think he is prepared to act if he thinks you are stealing? In that case the only flop you can be happy with has a 3 on it. Anything else and you are trying to cow him into folding. Given that he probably has something he likes I'm not sure this would be easy. I don't think you have the implied odds to carry on here.

In essence I fold because I don't like the prospect of the costly decisions I might have to make later in the dark. I'd make the raise, but I'd have wanted a fold or a call, not a reraise. I'm totally puzzled by your suggested rereraise. It won't make him fold there and then, gives him an invitation to set you all-in at best 50:50 if he fancies it, and it just makes the pot more costly for you if you lose it and may commit you on an awkward flop (555 for example).

I realise the "I wouldn't start from here" argument doesn't preclude me from answering the second part, but I have no answers to those at this stage.

Lurker

Re: Half an answer

[peterbirks.livejournal.com]

I have just noticed that I haven't put my own stats down here. Clearly this makes a difference.

I'm about 17.6%/13.5%, and my attempted steal on the button is about 36%.

Pete

[(Anonymous)]

The min re-raise with 33 made me a little bit sick in my mouth. I suppose it's purpose is to setup a steal on the flop so in all 3 cases, and indeed any flop whatsoever, I would bet half the pot and pray that he folds.

matt

[(Anonymous)]

Can I pre-emptively apologise for writing it's instead of its before you go off on a grammar rant.

[peterbirks.livejournal.com]

Yes, you may apologise.


Aand don't worry about it. I typo it's for its all the time. Fortunately, I usually notice it nefore it heads off into the outside world.


I tried to put down all options with no comment. I could have put a fourth option ("shove") or made this the third option instead of a mini-reraise. Of the two, I felt that the mini-reraise was fractionally less insane, which was why I chose it, but that was rushed. Perhaps it's the shove that is fractionally less insane.

PJ

[peterbirks.livejournal.com]

I'm not quite sure what you are saying here Matt. What's your preferred pre-flop response to Big blind's reraise?

Pete

[(Anonymous)]

I wouldn't reraise but if I had, then betting the flop seems mandatory if they check to you.

I would call as he hasn't really raised enough to deny you odds for hitting a set. But then I would fold the flop as it missed me.

matt

[peterbirks.livejournal.com]

Yes, I'm roughly with you on the "call/fold" line. So, we have a situation that crops up quite often, and yet two people would fold to the reraise and two people would call. (In fact, the actual hand which gave me the idea on this took a slightly different line, but I was looking to illustrate a general principle).

The research that has improved my bottom line the most in the past couple of months has been nothing to do with "sexy" hands and all-in decisions. It's simply been looking at what hands were profitable if I completed in the small blind to two or more limpers. I haven't seen that area mentioned that much anywhere. Three limpers to you and you have AJo in the small blind. What do you reckon?

PJ

[(Anonymous)]

"It's simply been looking at what hands were profitable if I completed in the small blind to two or more limpers."

Dangerous waters Pete. But of course if it improves the bottom line, it can't be all bad. Still keep in mind the signal to noise ratio of these type of analysis.

With 33, I'm calling the min raise, and folding all flops. Miniraising back is so strange line for me that I'm not qualified to answer.

Aksu

[peterbirks.livejournal.com]

Hi Aksu,

Well, it's obviously a bit more complicated than I wrote it. I suspect that I was losing with some hands that had the potential to be winners, but my play out of the blinds wasn't yet good enough. I've clearly also developed my style with hands that I am comfortable with, and have started steering away from hands which seem to get me into trouble or leave me a bit lost.

But it was also clear to me that I had to do something about my performance in the blinds, in that the (good) winning players I was up against seemed to perform about the same as me in non-blind positions (indeed, slightly worse), while they outperformed me in the Blinds.

There are a number of routes to travel when attempting to fix leaks that you can't really identify. Many players only look at "big pots", but I think that this is a mistake. I think that the leaks happen pre-fliop with hands that you don't really think about.

Perhaps I should fold hands where I call. Perhaps I should raise hands where I call. Perhaps I should change my level of raises. It's a matter of playing around with things and seeing if (a) you feel more comfortable and (b) your bottom line is getting better.

One counter-check suystem that is useful is that I can look at the performances at several sites. If I take A2o up to AKo on all these sites and compare my results to a default fold, then I can get a good idea as to trends.

What I discovered was that the average performance (for me) saw that limping in the small blind with multiple limpers before me was proving costly right the way up to ATo. AJo was marginal.

Now, one could argue that there's nothing wrong with the limp; it's just my post-flop play with it that sucks. Indeed, there is good evidence to support this, because my losses with these hands in the big blind were also bigger than I would hope.

I'm already thinking about how to play these hands in shorter-handed games, but you rarely get the limps round in those anyway.

I think that my main point was that a thoughtless completion with an Ace in the small blind (whereas you might discard 78o) might be costing you money. It's the small decisions that you make a lot that have the biggest impact on your ultimate profit (in the games that I play); not the massive hands where you stack people off or get stacked off in return. Mainly because stack-offs are so rare.

PJ

[(Anonymous)]

Those advocating folding 33 pre-flop are simply wrong. And it's not even close. I'd complete in the small blind with any ace and 78o after limpers. If you're losing money I would suggest the problem is with post-flop play and note that you need a stronger hand to play out of position.

matt

[peterbirks.livejournal.com]

Hi Matt:

My writing was poor there. In fact I do complete with 78o (down to 54o, in fact).

There's a clear reason why I get into trouble with these Ax hands post-flop (I had completed with them for some three months before I looked at the stats) and I know what that reason is. But I couldn't work out how to solve it (but I'm still working on it). So I decided, temporarily, to cut my losses and took the lesser loss of not completing. This saved me a LOT of money.

Looking at the offsuited connectors, I'm breaking even overall from completion in the small blind (i.e., performing as well as I would be if I just folded), but I would be well up if I hadn't been playing AKo incorrectly for a month or two(fucking FMM theories screwed me up there for a while).

Part of my problem is that I make so much more money when I put in a raise than if I complete (this applies in all other positions as well).

Here's some preflop stats and how things panned out:


No Raise: Won without showdown: 1,312 hands, net of plus $3655. $2.79 a hand.
Any Raise (by me): Won without showdown: 3,426 hands, plus $12,567. $3.67 a hand

No Raise: Went to showdown: 905 hands, plus $2,642: $2.92 a hand
Any Raise (by me): Went to showdown: 843 hands, plus $4290: $5.09 a hand

No Raise: Folded before showdown: 2,999 hands, minus $6,450: minus $2.15 a hand
Any Raise: Folded before Showdown: 1,190 hands, minus $10,194, minus $8.57 a hand


That gives a result of plus $5,455 for any raise, over 5,759 hands, at $0.95 a hand

While for No Raise I get a loss of about $100 over 5,200 hands.

This is over a sample of 41,500 hands on one site, btw, with $8,349 posted in blinds and a net loss of $2,220 in the big blind and $792 in the small blind. My win rate overall is 2.68 Big blinds per 100 hands.

I know that there's a serious problem with my post-flop play when I limp, because if I played percentages such as 16%/8% (a common line amongst many profitable players I am up against) I would be doing my bollocks. I know, I've tried it. So now I'm hammering away at something like 17%/14%, and that 3% where I don't raise just isn't winning for me at the moment.

PJ

[(Anonymous)]

Absolutely agree with Matt about folding 33 for minraise. The more likely the blind is having a monster, the more reason to call.

For me the no raise/any raise comparison looks meaningless. Could be a misunderstanding, but aren't the hand rankings in those groups quite different? On the other hand good stuff about identifying small pots the core of many problems, quite hard work to go through that kind of data.

Aksu

[(Anonymous)]

hehe, quote from Terrence Chan's blog:
"
50a/150/300. Folded to cutoff who raises to 850. I have KK in the SB. I pick up the wrong fucking chip and as a result, re-raise to just 1450. Cutoff obviously calls, and now his hand distribution is 100% of hands he originally raised with. :(
"

[peterbirks.livejournal.com]

Yes, it's a problem. In fact, with the hands ranked the same where I call sometimes and raise sometimes, I seem to be doing better with a raise. But here the sample size is smaller and I come up against your "signal to noise ratio" argument.

The numbers here are just backing up what i feel inside, that I perform much better when I have raised pre-flop and I am hammering away and know when to slow down, rather than when I have just limped.

I think part of it is that I seem to get a better handle on opponent's hand ranges when I am the aggressor. When I am the limper, I start thinking what I would do if I were the aggressor, but most of my opponents aren't thinking like that. So I end up misreading hand strengths and ranges.

PJ

[jellymillion.livejournal.com]

If you've been raising the guy's BB fairly regularly then I suppose he may have loosened his range a little and be letting you know he won't stand for it. Or he could (quite likely given the theme, I suppose) be 100% ABC and have a monster.

The trouble is that you only a long way ahead of 22 and hands with a 3 kicker. You're streets behind a pair and a small favourite ahead of two overcards. You're getting about 12-1 implied odds if you think you can double through him if you hit your set and he doesn't beat it. Without running it, it's quite likely to be +EV, but I don't expect by much. If the opposition is relatively poor then you probably don't need to court that much variance, the cash will arrive anyway.

I think I'm folding. If nothing else, it saves working through the flop options.