Off-Topic: pot odds, calling, and value betting

[dfscott]

Ok, I know this probably goes in another forum but like so many others, I feel more comfortable in this forum, so here goes.

I was trying to explain to a new player about value bets and how they relate to pot odds and deciding when to call. As I got into it, I realized I wasn't 100% sure which odds to use (1-card or "by the river" odds). Here's how I explained it -- please correct me if I'm wrong. And remember, I'm talking about mathematical probability here, not bluffing, semi-bluffing, slow-playing, etc.

A value bet is when you are betting a hand and hoping to get called because it is either:

1) the best hand (so you want to get money in because you expect to win it), or
2) it is a hand that has a chance to improve to be the best hand.

In the first case, you just bet regardless of pot odds. In the second case, you bet if your odds to make the best hand by the river are as good (or better) than the number of callers. For example, with the nuts on the flop, a value bet is always mathematically correct, but if you're 4 to the nut flush on an unpaired board (just under 2:1 to make your flush by the river), you need 2 callers to value bet.

On the other hand, if you're deciding whether to call a bet, you should only consider your chance of making your hand by the next card, not the river. (This is because if you don't make your hand, you might have to call another bet.) In the above example (the nut flush draw), you are just over 4.2:1 to make your hand on the next card. That means if it's $1 to call and you're heads-up against a single opponent (i.e., no value bet possible), you need $4.20 in the pot to call (I'm deliberably ignoring implied odds for simplicity).

As an aside, the nice thing about value betting is that pumping the pot can often make a call on the turn correct that wouldn't be otherwise.

Anyway, anyone disagree with this? Would anyone consider one-card rather than river odds for value betting?

[Brian]

The Theory of Poker by David Sklansky spells out in detail what exactly value betting is, as well as answers a few of the questions in your post. I suggest that you pick it up if you haven't already; it is the best book I have ever read on Poker. His section on hourly rate and expectation also helped me immensely with my tilt problem.

-Brian

[dfscott]

I've got TOP and read it a couple of times. I even pulled it out again just to look for this. However, I had a hard time finding anything about value betting. He talks about semi-bluffs, betting to drive people out, etc., but if value betting is in there, I couldn't find it.

As far a calling with correct pot odds, there's a lot about when to call with cards to come, but he takes a different approach on these calls. Sklansky says that you should consider any possible future bets in addition to this one. I'm trying to look at it from the other side, since it seems to be pretty difficult to figure out how many bets you'll be forced to call on later streets.

[AsusFull]

I thought it was 4 callers for a safe flush value bet.

I think if you're on the button during the flop and can buy a free card by betting or raising and get 2 callers, that would be a value bet.

But that's a fairly specific situation and you can't always guarantee they will check to you on the turn.

I dunno, that's how my n00b ass understands it. I don't think 2 callers for a flush on the flop is always a safe value bet.

[Zetack]

I think value bet is term that has a more specific connotation than simply +Ev betting. Usually (not always, but usually) it refers to a bet on the river.

I think of it as a crying bet, the equivalent of a crying call. The hand is either mediocre, or the action or the board makes it look like it may not be the best hand, but you judge there is sufficient likelihood hood that it is the best that you bet anyway--for value. A classic example is betting the river with your top pair when the flush card hits.

--Zetack

[Nak]

When you are betting for value, it means that you are making money on your bet. For example, on a flush draw (assume it's the nuts, etc.) if the pot is $10 and you can bet at it for $1 you are getting 10:1 on your flush draw there. You are making money if you bet - hence the term "value".

Nak

[dfscott]

Ok, AsusFull got the closest to what I was looking for, but I don't think I was clear in my original post. I'm not looking for a definition of a value bet -- I'm looking for the way you determine when a value bet is profitable when you're on a draw.

We all know it's based on a) the money you're putting in, b) the money that is (or will be) in the pot, and c) the odds you have of hitting your hand.

a & b are pretty straightforward. The question is: what is c? Is it the odds you'll hit your hand on the next card or the odds you'll hit it by the river?

I have always used the following rule, but now I'm not 100% sure why:

1) To call a bet, you must be given sufficient pot odds based on your likelyhood of hitting you hand on the next card. This is because if you don't, you'll likely have to call another bet on the next street.

2) To make a bet, you must have enough callers so that you're getting sufficient odds to hit your hand by the river. This is where it gets little fuzzy for me, but I think it's because you're planning on staying in, so you want to get the bets in.

I'm beginning to think that maybe I'm putting to much emphasis on this and should just teach good play, but I think that it's important to ground people in fundamentals before they start calling based on a "feeling." /images/graemlins/smile.gif

[easypete]

[ QUOTE ]
I thought it was 4 callers for a safe flush value bet.

[/ QUOTE ]

This is for a call to see one more card. If you are 4 flushed on the flop, you still have 2 cards to catch one of the 9 remaining flush cards.

9/47+9/46 = 0.39.

So if you have 2 opponents to call a flop bet, you are making money (in the long run). Your 1 bet called by 2 players is giving you 2:1 odds to make a 1.6:1 shot by the river. So if you can cap it on the flop with more than 2 players in the pot, you are making some increadible theoretical money.

Now on the turn, if you don't make your card. You only have one card to come.

9/46 = 0.20. So you are 4.1:1 dog. If you have 2 oppenents in the pot, a bet now is a semi-bluff. You are trying to win the pot, but the bet is now -EV (assuming you get called). Now if you have 5 oppenents at the turn, a bet is a value bet if all players call (assuming a huge pot and you have the lead) because you are getting more on your bet than your chances of winning with one card to come.

So one way to takle this is discussed in many 2+2 books. If you are in an EP and are willing to call w/ a draw, then bet it. You may win the pot right there, and you may improve to the best hand.

[spamuell]

[ QUOTE ]

When you are betting for value, it means that you are making money on your bet. For example, on a flush draw (assume it's the nuts, etc.) if the pot is $10 and you can bet at it for $1 you are getting 10:1 on your flush draw there. You are making money if you bet - hence the term "value".

[/ QUOTE ]

This is wrong. If the pot is already $10 then although it is +EV to put $1 in the pot and draw to a flush, this is not a value bet as it would be even more +EV to put $0 in the pot and draw to your flush. It is a value bet when you are making more money by putting in more money. For example, if you have a flush draw on the flop and you are getting 10 callers (assuming a flush is the nuts etc), it is massively +EV to put as much money as you possibly can in the pot. It is more +EV to put $1 in the pot than $0 and even more +EV to put $10 in the pot (assuming that you will still be called). This is true because you make your flush by the river about 35% of the time so, essentially, 35% of the money in the pot is yours. Put another way, you are about 2:1 (1.86:1) to make your flush by the river, but you are getting 10:1 on your bet.

Thus, a value bet only refers to any money that is being put into the pot on this betting round in order to match the money you are putting in. The size of the pot is irrelevant (except indirectly because you might be more likely to be able to make a value bet because a large pot could induce a call).

[Nottom]

The best definition of a value bet I can think of is that it is a bet that makes money when called. What this means is that if you take the same situation and make that bet 1,000,000 times you will make money.

Esentially all you said is correct, so I think you've got it straight. The only time I can think of that you would really worry about one card odds with value betting is if you have a really big draw on the turn and are considering you options.

[Nottom]

[ QUOTE ]

I dunno, that's how my n00b ass understands it. I don't think 2 callers for a flush on the flop is always a safe value bet.

[/ QUOTE ]

You are 1.86:1 to make you flush by the river, if you are getting 2:1 on a 1.86:1 proposition there is definate value in the bet.

[Nottom]

[ QUOTE ]
When you are betting for value, it means that you are making money on your bet. For example, on a flush draw (assume it's the nuts, etc.) if the pot is $10 and you can bet at it for $1 you are getting 10:1 on your flush draw there. You are making money if you bet - hence the term "value".


[/ QUOTE ]

This is completely wrong and is a common misconception. The size of the pot is practically irrelevent when it comes to value bets.

[Nottom]

[ QUOTE ]
2) To make a bet, you must have enough callers so that you're getting sufficient odds to hit your hand by the river. This is where it gets little fuzzy for me, but I think it's because you're planning on staying in, so you want to get the bets in.

[/ QUOTE ]

This is correct, you are gonna still be around on the river with pretty much any draw big enouhg to worry about value betting, so you use the odds of making it by that point. Notice that if you are playing No-Limit then all this value bet talk is basically garbage unless you have a true monster draw (14+ outs) that would actually make you a favorite to complete by the river.

[spamuell]

You are correct in your explanation of being able to make a value bet with two cards to come because you will not be folding by the river but your calculations are slightly off:

[ QUOTE ]
This is for a call to see one more card. If you are 4 flushed on the flop, you still have 2 cards to catch one of the 9 remaining flush cards.

9/47+9/46 = 0.39.

So if you have 2 opponents to call a flop bet, you are making money (in the long run). Your 1 bet called by 2 players is giving you 2:1 odds to make a 1.6:1 shot by the river.

[/ QUOTE ]

The actual figure is 1.86:1. Your error comes from the calculation 9/47 + 9/46. Check out this (http://forumserver.twoplustwo.com/showflat.php?Cat=&Number=517275&page=0&view=expand ed&sb=5&o=14&fpart=1) thread with an excellent explanation by Nottom of how to correctly work out these odds.

[chief444]

dfscott,

This all sounds pretty good. The one thing I noticed that I don't really agree with is this:

[ QUOTE ]
As an aside, the nice thing about value betting is that pumping the pot can often make a call on the turn correct that wouldn't be otherwise.


[/ QUOTE ]

If you would not have even had odds to call the turn then it is likely that you're value bet on the flop was not a good one. In extreme situations (I remember a straight flush draw vs. made straight posted a few months ago in the probability forum that is an example) this may be correct but even then it's close to 0 EV. The only reason I point this out is I often see someone posting that they bet or raised early in order to "give themselves the right pot odds to call later". This is not good thinking. Raising until it's capped on the flop with a very weak draw or bottom pair will probably give you correct odds to call the turn but that doesn't mean it's a profitable play. I realize that isn't even what you meant but just wanted to try to clarify somewhat. In general I agree with what you are saying in that you usually value bet strong draws based on two cards to come when you know you will have the odds to see both cards anyway.

[Nak]

[ QUOTE ]
[ QUOTE ]

When you are betting for value, it means that you are making money on your bet. For example, on a flush draw (assume it's the nuts, etc.) if the pot is $10 and you can bet at it for $1 you are getting 10:1 on your flush draw there. You are making money if you bet - hence the term "value".

[/ QUOTE ]

This is wrong. If the pot is already $10 then although it is +EV to put $1 in the pot and draw to a flush, this is not a value bet as it would be even more +EV to put $0 in the pot and draw to your flush. It is a value bet when you are making more money by putting in more money. For example, if you have a flush draw on the flop and you are getting 10 callers (assuming a flush is the nuts etc), it is massively +EV to put as much money as you possibly can in the pot. It is more +EV to put $1 in the pot than $0 and even more +EV to put $10 in the pot (assuming that you will still be called). This is true because you make your flush by the river about 35% of the time so, essentially, 35% of the money in the pot is yours. Put another way, you are about 2:1 (1.86:1) to make your flush by the river, but you are getting 10:1 on your bet.

Thus, a value bet only refers to any money that is being put into the pot on this betting round in order to match the money you are putting in. The size of the pot is irrelevant (except indirectly because you might be more likely to be able to make a value bet because a large pot could induce a call).

[/ QUOTE ]

Yes, I understand what you are saying. What I mean is that if I already have odds for my flush draw, then any callers that I get will earn money with a value bet. Whereas if I don't have pot odds already then I need to consider implied odds etc.

Thanks for your clarification.

Nak

[chief444]

[ QUOTE ]
What I mean is that if I already have odds for my flush draw, then any callers that I get will earn money with a value bet. Whereas if I don't have pot odds already then I need to consider implied odds etc.


[/ QUOTE ]
Just because you have the pot odds to call does not mean putting more money in the pot is +EV for you.

[dfscott]

[ QUOTE ]
dfscott,

This all sounds pretty good. The one thing I noticed that I don't really agree with is this:

[ QUOTE ]
As an aside, the nice thing about value betting is that pumping the pot can often make a call on the turn correct that wouldn't be otherwise.


[/ QUOTE ]

If you would not have even had odds to call the turn then it is likely that you're value bet on the flop was not a good one. In extreme situations (I remember a straight flush draw vs. made straight posted a few months ago in the probability forum that is an example) this may be correct but even then it's close to 0 EV. The only reason I point this out is I often see someone posting that they bet or raised early in order to "give themselves the right pot odds to call later". This is not good thinking. Raising until it's capped on the flop with a very weak draw or bottom pair will probably give you correct odds to call the turn but that doesn't mean it's a profitable play. I realize that isn't even what you meant but just wanted to try to clarify somewhat. In general I agree with what you are saying in that you usually value bet strong draws based on two cards to come when you know you will have the odds to see both cards anyway.

[/ QUOTE ]

Actually, I think I agree with you here. I tried to work out the math and couldn't find any reasonable situations where calling the turn would be contingent on betting the flop.