On Folding the River

[W. Deranged]

This is something I've been thinking about recently.

We'll start with a scenario:

Villain is in the MP2 with 88 and open-raises. A loosish but moderately aggressive villain cold-calls in the CO (he's 27/9/1.7). The blinds fold and it's two of you to the flop:

Flop: Q/images/graemlins/heart.gif T/images/graemlins/club.gif 3/images/graemlins/club.gif

You bet, and your opponent calls.

Turn: 7/images/graemlins/heart.gif

You bet, and your opponent calls.

River: K/images/graemlins/diamond.gif

You check, and your opponent bets...


In many ways this seems like a pretty easy fold. Villain has called you on two streets. He cold-called a raise pre-flop. Now the K hits and he bets the river with confidence.

But let's consider this another way.

If you were villain, what sorts of hands would you bet on this river if checked to? Would you bet a hand like JTs? QJs? AT? 99?

If we think about the traits that are most common among standard opponents at the small stakes on the river, two things come to my mind for a situation like this:

1. Our opponents generally fail to value bet WAY too much.
2. Our opponents seem to like to bet when checked to when they don't have much, hoping you'll fold.


Basically, it seems to me that in general, our opponents are far more fearless on the river when they have NOTHING, then when they have a moderately good hand that is probably good. (This is neglecting situations where they have very strong hands). Our opponents HATE getting raised on the river, and so will often fail to bet hands that we 2+2ers have trained ourselves to value bet. Yet they'll often attempt bluffs that we probably wouldn't find value in.


So my conclusion is that, in the situation above, our opponent is most likely to be betting one of the following two types of hands: A very strong hand (KQ, KT, J9, TT, and so on), or a very weak hand (66, A/images/graemlins/heart.gifx/images/graemlins/heart.gif, A/images/graemlins/club.gifx/images/graemlins/club.gif, and so on). He's much LESS likely to be betting a hand like QJ, JT, AT, maybe even K9 or KJ, and so forth.

The consequence of this is that, when considering how our hands stack up against villain's range on the river, we need to "terraform" that range, significantly depressing the likelihood of a lot of "middle-range" hands. So the shape of villain's range is kind of bimodal, with peaks on the bluffing end and strong hand end, and a depression in the middle. If we look at how our hands stack up against that range, I think we'll find that in fact a hand like 88 is in far BETTER shape than might immediately be apparent.

So in situations where we've encouraged villain to bet a lot of hands, even if it seems that there are TONS of hands that beat us, maybe we should be calling more. A lot of those hands that are beating us are hands that the vast majority of small stakes opponents simply don't like to bet.

[Nick Royale]

Nice post. I don't know to what extent this theory is applicable, but the idea certainly has a point.

[Jake (The Snake)]

Hey Deranged,

This is a concept that is also at play when Hero is in position with a decent hand and villain randomly donks the river. Since he usually has either a really good hand or a complete bluff, there is often not much reason to raise. Maybe an example (http://forumserver.twoplustwo.com/showflat.php?Cat=0&Number=4175994&page=0&fpart=1&v c=1)?

By the way, I can't remember who it was, but somebody used to talk about this concept quite a bit (I think maybe Joe Tall?). Anyway, it's something that I think is quite applicable to many river decisions.

Good post.

[hobbsmann]

Knowing what to do on rivers like this is trait I've picked up from playing a lot of 6 max. So often if I have a hand with any kind of showdown value (the above hand is a great example) and there is an obvious draw that doesn't get there on the river AND my hand is fairly weak, checking with the intention of calling a bet is often the best play.

Value betting this river is most definately wrong as villain's range of calling hands is going to be ahead of us, but by checking we allow the villain to bluff with a wide range thus putting ourselves in a EV+ situation. You should be less inclined to check to induce a bluff against a horrible player as they will call with worse hands and you should be ready to check/fold against a solid player who will not bluff/value bet with worse hands.

Very nice post and thoughts Deranged. Interesting theory and something I will try to keep in mind as I play.

Nice post.

[WillMagic]

[ QUOTE ]
Hey Deranged,

This is a concept that is also at play when Hero is in position with a decent hand and villain randomly donks the river. Since he usually has either a really good hand or a complete bluff, there is often not much reason to raise. Maybe an example (http://forumserver.twoplustwo.com/showflat.php?Cat=0&Number=4175994&page=0&fpart=1&v c=1)?

[/ QUOTE ]

It's funny...I think of it at another level entirely. Most players think like you do and are really, reluctant to raise river donks, but entirely happy to call them with as ace-high, because they see them as bluffs/nuts. So often I find myself bet/folding random rivers OOP into people with weak pairs, when I think their range is something like AK-AJ unimproved and decent top pair/mid pp hands. It doesn't even matter that they've had the lead the whole way.

Will

[silkyslim]

i like value betting this against our favorite opponents, but against aggressive opponents like this i will check/call if the pot is big enough. Ive won with 66-99 alot with this kind of situation (warning: small sample size) . What sparks you to think about such things?? I probably do but neglect to write it down. does anyone keep a poker journal?? maybe with a little locket and a pink pony cover??.........

[Jake (The Snake)]

[ QUOTE ]
I can't remember who it was, but somebody used to talk about this concept quite a bit (I think maybe Joe Tall?).

[/ QUOTE ]

Just remembered it was Josh. (sthief09).

[toss]

Excellent post. That means we should adjust our play and checkcall more often than what would seem right. I think.

[chief444]

Nice post. I agree with everything except that it seems like an easy fold...for reasons you already stated.

[Harv72b]

Let's take this a step further, shall we? /images/graemlins/grin.gif

If you're going to take this line and call the river bet from time to time with a weak holding, then you also need to bet the river from time to time with the same weak holding. And likewise, you need to check/raise the river with a strong holding every so often. Basically, you want to strike the fear of God into anyone who enters a pot against you, because you never want to be predictable in your play. Especially not HU in the higher small stakes/mid stakes limits.

In this scenario, the flop brings two obvious draws which the villain might call down with and then bluff at the river with if checked to (although this is a pretty bad board to bluff against a PFR on). He could have J9s (a 27 VPIP guy will definitely coldcall with that hand, particularly if he's shown some semblence of positional awareness) or KJ for the OESD. He could have two clubs for the flush draw. The problem here is that only the flush draw, assuming he didn't pair a Q or K on the way, is behind you after the K hits on the river. So you're basically check/calling here in the hopes that he's going to bet either a busted flush draw or a worse pocket pair. While I agree with what you've said about opponents' willingness to bet weak hands vs. borderline ones on the river, I don't think it happens all that often on this particular board vs. that opponent. I think you can pretty safely check/fold this one, getting only ~6.5:1 on the river call. If the pot or your pair were a little bigger (say JJ), then I'd be a bit more willing to check/call.

Note that you are also hoping that this villain will be aggressive enough to bet the river with a busted flush draw after being passive enough not to raise the flop with the same draw and position.