Wednesday, February 08, 2012

The min raise

Last night I was playing at the Golden Nugget with the visiting Lightning36. He lives in my hometown and works at one of the colleges I attended, so it's always good to catch up a bit with him when he's in town.

So I have Ah-Ks in early position and raise to $8. Three people call me. The flop is all hearts, Q-8-3. With the nut flush draw, I like how this looks. I bet $20, about 2/3 of the pot. One fold, then another player raises to $40. Another fold and it's back to me, heads-up.

The pot is about $90, and it costs me only $20 to call, giving me 4.5:1 pot odds. There are nine cards that the dealer can find in the deck to make my flush, and 47 unknown cards, so the probability of hitting on the turn is 9/47, or 19%, which is 5.2:1. Even if I have no implied odds--i.e., my opponent won't put in another dime if the flush comes and it scares him--I have favorable odds to make this call correct.

As soon as I have put in another $20, the dealer peels off the 4 of hearts. Nutterific! I had a feeling that that was going to be an action-killer, because his raising range probably did not include any hands with which he would have liked to see a fourth heart come. I was right--he checked it back to me.* That pretty much sealed the conclusion that he hated that card.

Fifth street was a blank. I led out for $55 into the $110 pot, hoping that it was small enough to tempt a call. But he thought long and hard, in apparent agony, before finally mucking face-up his pocket 8s (flopped set).

And thus we see the chief problem with the minimum raise: It almost never accomplishes whatever purpose the raise had.

Here he presumably wanted me to fold, but the size of his raise actually meant that a fold would have been mathematically incorrect.

Let's consider some other reasons that one raises.

1. As a bluff, when you sense weakness. But a min-raise also looks weak, and may just induce an over-the-top re-bluff.

2. To price out draws. But as this example shows, a min-raise often fails to accomplish that, because it sets the price of the draw temptingly low.

3. To build the pot when you have the best hand. But if that's your goal, why be so chintzy? Put in some serious money instead and hope for a call.

4. As a probe, to "see where you are." But a min-raise often doesn't clarify the situation for you. In this case, a fold would have told him that I completely missed the flop and had nothing, but my call left him in the dark as to whether I had an overpair with no heart, or was drawing to the flush. That uncertainty is what gave him the agonizing decision when I put him to the test on the river. Had he made his raise more substantial, the first advantage is that I might have given up the hand, knowing that he was unlikely to stack off if I hit. The second advantage is that he could be more confident that I was not calling with, say, pocket kings with no heart, making his later decision much more clear.

Like any poker move, the min-raise has a place. But it's really a very small niche. It's a specialty tool to be used in a few specific, rare situations. For example, you have a monster hand, know that an opponent is weak, and a minimum raise is the most you can possibly hope to squeeze out of him. Or you have an overly aggressive opponent whom you have reason to believe will read a min-raise as weak, when you actually have him crushed; the min-raise in that situation is a trap that you hope will induce a reraise that will get him pot-committed in bad shape. Or you have a lock on the hand and a whole raft of callers to somebody's initial bet; a min-raise may be the best way to swell the pot by giving every single one of those callers temptation to put more chips in when they're drawing dead.

How often do situations arise in which the min-raise is the optimal move? In my view, it's rare. I doubt I put in a min-raise more than a couple of times a month.

Imagine that you've produced a gadget that you want to sell, and you have to determine the price. If you set it too low, you'll sell a ton, but you might actually lose money if revenues are below your costs. If you set the price too high, you'll make a handsome profit with each sale, but sales will be slower than they would be if the price were lower. There is a perfect price somewhere in the middle that optimizes your profit, at the top of a bell-shaped curve of price versus profit. Finding that sweet spot is the key. Large corporations have whole divisions of financial wizards whose only job is finding optimal prices for their products.

When you bet and raise in poker, you are, in effect, offering your hand for sale. You need to set the price to maximize profit. Sometimes the way to do that is by going all-in, if you think your opponent has a second-best hand and can't get away from it, or is unusually loose and calls anything. But if you made every raise all-in, you'd lose your shirt, because most of the time worse hands won't call you and you'll make zero, and once in a while you'll get called by a better hand and lose everything. So that should not be your default or most usual raise strategy.

Similarly, if you habitually min-raise, you're setting your price much too low. Sure, you'll get called a lot--but that includes spots where you really don't want a call, and it leaves money on the table in most of the spots where you do want a call.

For any raise you make, you need to have in mind a clear purpose. What are your goals? What better hands do you think will fold? What worse hands do you think will call? Most importantly, what raise amount will make the most money? I think you'll find that the smallest raise is almost never the right tool.

Min-raises do not lead to max-profits.

Oh, about that image above: It's Min, an Egyptian god of fertility. I did an image search for "min," having no idea what I might find. When I saw that phallic image, and paired it in my mind with the concept of "raise," well, my 12-year-old self took control and decided that had to be the illustration of the day. It is, quite literally, the Min raise.

*One could certainly argue for a lead-out bet on this turn. After all, I have a big hand and want a big pot. The obvious, straightforward move is most often the right one, and putting more money into a juicy pot when one has the best possible hand is certainly the obvious play. My instantaneous analysis, however, was that calling the flop raise and then leading out was too transparent for having the flush. My plan was to let him bet at it again and check-raise; if he didn't bet, then having passed on the turn, a bet on the river might look like I'm working a non-flush overpair. That is, if there is X amount that he's going to be willing to call, he might be more prone to do so on the river after a check-check turn than immediately after the scare card hits. That was my thinking at the time.

However, there is this strong counter-argument: If he has a set or two pair, he will actually be more willing to put in his call of X amount on the turn than on the river, because he can still hope to improve to a full house, a possibility that is foreclosed after the last card is on the board. Furthermore, if the board does pair on the river, I'm suddenly in a very awkward spot. I will have missed my chance to win the pot when I had the best of it, and will kick myself for giving him the free card. In retrospect, I'm inclined to believe that leading out on the turn probably would have been better, but I think the pros and cons come out close enough that if it was a mistake, it was a relatively small one.


PokerLawyer said...

This is terrific, and I'm really glad I read it. Thanks for writing it, grump.

sevencard2003 said...

why not reraise allin on the flop so if the heart does NOT come on the turn u wont get priced out of getting to see both cards

Rakewell said...

Well, I guess that's one possibility. However, a raise sequence of 20 -> 40 -> 200 is pretty unusual.

More importantly, though, I think my fold equity is near zero. The hands with which he is raising me on a one-suit flop are limited to flopped flushes, sets, and maybe two pair. I don't think he'll fold any of those. Without any meaningful fold equity, I have to hit in order to win, and I'm only about 28% to beat either a set or a small flush. It's OK to make that kind of power play when you have a decent chance of folding him out, but since that's unlikely here, I don't think it's a good move--it would be getting all the money in as a 3:1 underdog.

lightning36 said...

That was a fun hand to watch. As I think I told you, I correctly (for once - ha!) put you on A-K with either the A or K of hearts.

I must admit, though, that I liked playing with you much better when I had position on you. : o )

Rakewell said...

Hey, I gave you the option to be on my immediate left. You declined. :-)

BTW, I do this with nearly all out-of-town visitors. When I go to the trouble of playing at the same table for ease of chatting, I'd rather do it in close proximity than across the table. So I request a seat change button and move when an spot adjacent to the friend opens up. But if the one that becomes available is the one of the left, I don't want to be seen as grabbing the tactical advantage, so as a courtesy I routinely offer to let the other person shift to the left and I'll sit on the right. I'm OK with it either way.

Michael (@jinxclev) said...

Great analysis. I'm also glad there aren't many/any in my poker room that read your blog, as they are very fond of the min raise, and it's one reason I think I'm able to extract more value.

Anonymous said...

Your turn check actually made your flop call incorrect against his hand and strategy (of always checking/folding any unpaired 4-flush board) - if he's going to force you out of the pot on a blank turn, you only win (9/45)*10/44=15.45% of the time.

Of course, that's assuming he'd bet the turn big enough to make you fold.

Anonymous said...

Sir I read and enjoy your blog regularly and I am sure it was just a big figure typo when you calculated 19% to be 5.2:1 instead of 4.2:1 (9/47)

Take care

Wine Guy said...

Another great article, but I have a question that is always in the back of my mind..If you were in the position he was in, what, to you, would have been a valid raise? it is always disconcerting to me when I raise and half the table stays in. I don't wish to make an over bet, but to me this is the most difficult part of my game that needs attention..Thanks,..

Anonymous said...

been playing for near 25 years. I laugh my ass off when I see a guy do the min raise and get mad he gets called. I might have done it 10 times since I have started playing. This is why we can win money and others are always trying to figure out why they lose.

Rakewell said...

Wine Guy: I suppose something like 3x-3.5x my bet would be about right. That guarantees that I'm paying too much to see the next card, if he is capable of folding and thus not giving me any implied odds for stacking him.

Wine Guy said...

Thanks for the insight..Appreciate your information!