a Question for the Expert (Ed Miller)

[Mikey]

Here is a good question referring to finding hidden outs located on pages 108 and 109.

Let us assume the same exact hand and the same exact scenario but that now the river card does improve your hand which is basically any J or any card that pairs the board.


What will you do now? Raise? or Call.

Let us assume that you will only call if any of those cards that improve your hand actually do help you.

Then you're effective odds that the pot is laying you on the turn if you do decide to call this bet and one on the river is or we can approximate to (and this all depends on the range that the BB fold right now when we make the call on the turn or call and pay off all the way to the river as well.)

Our effective odds right now are in the neighborhood of
14 to 2 or simply 7 to 1 if the BB drops out immediately or as high as 8 to 1 if he simply calls now and on the river.

NOW this is assuming we are calling the river if we improve and not raising if we improve, because if we raise when we improve to any of these outs, our effective odds drop considerably and this is assuming the BB folds to the river raise but calls the turn bet to approx. 6 to 1 if the original checkraiser just calls our river raise and if he reraises our raise and we pay off, it is approx. 5.5 to 1.

These are all the effective odd answers I'm coming up with with my calculations. I hope you understanding how I'm coming up with these numbers. NOW........

5 outs with 1 card to come is 8.2 to 1. Why would I consider calling here when my odds that pot is offering me is effectively in the range from 8 to 1 and the range from 5.5 to 1.

Thanks in Advance.

*Mikey

[DonkeyKong]

maybe I am misunderstanding your post but it looks to me like you are mixing up the betting rounds.

you have to separate each round of betting with the associated odds.

you have 5 outs so you need 8.2 to 1 to call. you are getting 12:1 here on your call of the raise so that is good. you can't exactly calculate your implied odds because that is in the future... you can estimate it but that doesn't matter that much in this case... the fact is that you are getting the right odds to call right now. if you don't improve, you won't bet again but you may win more bets if you do improve so your implied odds are equal to or greater than the pot odds.

You seem to be saying that if after you make your hand (you hit your out), you are then going backward in time and calculating the odds back on 4th street -- if I am understanding you right. I think
you need to seperate each round of betting.
If you make your hand on the river, now this is a new scenario. You may only get paid 1:1 on your river bet (you bet, 1 caller) but that shouldn't negatively affect your 4th street bet.

your pot equity is much higher on the river becaues you have a made hand. Let's say you think you are best so you are 75% to win the pot (the other guy might have a set)... but before, your pot equity was only about 11-12%. The value of your hand just went up ~7x. when you will win 75% of those situations; now you are 'betting for value.'

If a Jack comes, a check/raise might be correct as he can't expect that a Jack helped you in this case.

--------------
EDIT: I might not be getting my implied odds comments right but I do think you are making it harder than it is...
You don't need to calculate implied odds, you just have to know when implied odds are better than pot odds and when you have reverse implied odds (redraws etc...).
--------------

Trent, Mikey and Sue (if you can get that reference, you da man)

[Mikey]

Ok... first of all.

You have position on both opp.

On the turn you have been checked to by both opp.
ACTION: You bet, BB calls, and not Middle guy checkraises.

The river now is definitley a new scenario and it is my best estimate from experience that he who checkraises the turn almost always bets the river, so effectively you will be putting in 2 BB to see this hand through that is why I'm calc. effective odds since you are going to also put in an extra bet on the river.

You will be effectively putting in 2 BB to see this hand through if you improve therefore I'm adjusting my odds to the pot size and the future bets I will be putting in.

This type of hand is not the same if you are drawing to a straight or a flush or any type of lock hand. Those hands are easy to calc. effective odds at that particular interval, because if you miss your straight, you won't be putting in any more bets, if you miss your flush, you won't be putting in any more bets.

This hand is a little different, because we aren't sure if when we hit any of those cards if we have a win, we may or may not, therefore I think it should be calculated differently.


Trent, Mikey and Sue (if you can get that reference, you da man) *SWINGERS

[Redeye]

You cannot count the bet you put in on the river when deciding what odd the pot is laying you. In your calculation on the turn, you are assuming that the hand you are drawing to will be a winner a large portion of the time. If there was little chance the hand your drawing to would be the winner at showdown, you would instead need to discount the outs you are counting and recompute your pot odds. On the river, the bet you put in is because you think you have the best hand and is a value bet.

[DonkeyKong]

Mikey, you have a point in your post that is probably worth discussing...

in all likelihood, and the crux of the Miller text, is that you will always call bets when the pot is big so long as you have some small chance that you are not dead. (the Miller text also suggests it is sometimes right to call even if you strongly suspect you are drawing dead -- that is, in a huge pot).

but in this case, you have top pair. the chance is not ZERO right now that you are ahead. you have to think it reasonable that you have some live outs PLUS the very small chance that you are ahead.

if you don't hit one of your live outs, it still might be correct to call based on the chance that your opponent is bluffing or he semi-bluffed and didn't get there. this seems to be more your point... Since the pot is big, you know you are going to call the river bet if you don't hit your out so you might as well factor that in on 4th street.
I kind of agree with you here and it seems inconsistent to do otherwise but given the fact that you just can't lay down every time you get raised (others will start doing this to you), pay 1 extra BB and keep the odds calculations intact...

besides, you never know what kind of crazy stuff people will take a shot at you with. I had a guy re-raise me tonight with.... 5th pair (his pocket 6's with a 7 Q K x T) vs my pair of kings-ace kicker... I thought he had a straight for sure... looked him up b/c it was only 1 extra BB in a good pot... wtf?

thanks Ed

[Mikey]

"besides, you never know what kind of crazy stuff people will take a shot at you with. I had a guy re-raise me tonight with.... 5th pair (his pocket 6's with a 7 Q K x T) vs my pair of kings-ace kicker... I thought he had a straight for sure... looked him up b/c it was only 1 extra BB in a good pot... wtf?"

I agree with you here because you are heads up, but the opponnent we are describing in my situation is check raising not 1 but 2 players and is voluntarity putting money in the pot out of position as well as betting the flop on an uncoordinated raggedy flop board.

[Mikey]

"You cannot count the bet you put in on the river when deciding what odd the pot is laying you."

Effectively you can and I think you have to because there is a final betting round in which your opponent will almost always bet.

"On the river, the bet you put in is because you think you have the best hand and is a value bet."

You aren't value betting anything unless your opponent checks to you. You have just been checkraised on the turn and on the river you will probably be just calling as well because your opponent will be betting the river almost all of the time.

calling twice = 2 BB


Like I said before you are not drawing to a made hand, the hand you are drawing to may not even be good if it hits. You aren't drawing to 4 outs where if you hit the out you have the nuts are near nuts. The hand you are drawing to may not even be good if it hits.

[DonkeyKong]

I agree with everything but:
-----
Like I said before you are not drawing to a made hand, the hand you are drawing to may not even be good if it hits. You aren't drawing to 4 outs where if you hit the out you have the nuts are near nuts. The hand you are drawing to may not even be good if it hits.
-------

You have already adjusted the outs to compensate you... you have adjusted them down. if it doesn't hit, it doesn't hit but the pot odds say go for it.

[Mikey]

Range from 8 to 1 to 3.4 to 1.

This is how I came up with these numbers.



BB folds as you complete the action on turn
Checkraiser bets and you call on river.

Effective odds = 6.5 to 1



BB calls and completes the turn action.
BB checks. Checkraiser bets. You call. BB folds.

Effective odds: 7 to 1.



BB calls and completes turn action.
BB checks. Checkraiser bets. You raise. BB folds. Checkraiser calls.

Effective odds: 5 to 1.



BB calls and completes turn action.
BB checks. Checkraiser bets. You raise. BB folds.
Checkraiser caps it. You call.

Effective odds: 3.4 to 1

[Guy McSucker]

Mikey,

You're onto something here, for sure. You should certainly consider the future action and effective odds in these situations, because when you call on the turn, you are committing to calling on the river some of the time, and you won't always win when you do.

(You rightly pointed out that this is why these situations differ from drawing to the nuts.)

It's quite tricky to do these calculations precisely, because you won't be calling all the time: you only call the river when certain cards hit. Here's my take on how to think about these things.

I don't recall the example exactly, but let's suppose there are ten cards that might improve you to a winner, of which five actually will improve you to a winner, but you don't know which five.

Your game plan is to call now, then call on the river if one of the ten possible good cards hits.

Consider the river bet alone. You call it when any of your ten good cards comes. Half the time, you win, because it was one of your five genuine outs. Half the time, you lose, because it wasn't.

You therefore break even on the river against one opponent: you're getting even money on an even money shot. (You might even make money on the river against more than one foe, but perhaps not so many of your outs are clean against two opponents.)

In this case, since the river action is break-even, you can ignore it for the purposes of your effective odds calculations, so you just look at the pot odds on the turn as if you had five known clean outs. How convenient!

Suppose instead your five outs are spread across fifteen potential cards. You call on the river when one if these 15 cards hits, but only win when one of the five good ones comes. Your river call loses 2/3 of the time and wins 1/3 of the time, so on average you lose 1/3 of a big bet on the river.

Now you need to look at the pot odds on the turn, and see if your "five outs" gives you an overlay of more than 1/3 of a big bet.

In this case you're getting 12-1 or so, on an 8-1 shot. With five outs, the EV of the turn bet is

(12 * 1/9) - (1 * 8/9) = 4/9 which is more than 1/3 of a big bet, so you have a call, just.

In reality things aren't this exact of course. Sometimes he won't bet the river, so you don't have that loss of 1/3 of a bet; sometimes your outs estimates are wrong, so you lose more often than you expected; sometimes you're in front anyway!

That's just poker. You have to make do with your best guess. Once you're good at this kind of thing, the various inaccuracies tend to even each other out, so you can get away with calculating as above.

Guy.

[Ed Miller]

I don't have time to address this question thoroughly. Here's my "brief" take on it:

First of all, this was the first thing Barry Tannenbaum mentioned to me about my book. He said, "I have a problem with your Hidden Outs section," and then he basically made your point, Mikey.

You (and he) are partially correct. There is indeed betting on the river, and that betting can change your effective odds in this situation.

But where he is wrong (he basically thinks that this issue invalidates the reasoning in my section... and you should actually be folding a lot of hands that my analysis would have you call with), IMO, and where you may be wrong (I haven't read all your stuff in this thread) is that I think, in general, the river action roughly balances out. That is, you end up paying off when you "improve" with a card that wasn't actually an out roughly as often as you actually improve and win an extra bet on the river.

Now obviously I can't enumerate every scenario and prove it conclusively, but I can say with relative certainty that the effect is somewhat small. That is, if you calculate your hand as having 5 outs, and your pot odds make it slightly correct to call with those 5 outs, the river action can change your effective odds only slightly (because of the balancing effect)... making it at worst only very slightly wrong to call with your 5 out hand.

The reason I'm not worried about that point is that the whole process for determining the 5 outs in the first place is fraught with guesses and uncertainty. After all, you have to guess about what your opponent might have. So the small adjustment needed for the river betting is already within the fairly large margin of error. Thus, it is negligible.

BTW, I thought this through before I wrote the Hidden Outs section... and I made a conscious decision not to mention this wrinkle at all. I don't think considering it can help you make better actual poker decisions.

So basically, I think I agree with Guy. /images/graemlins/smile.gif

[Mikey]

But what happens on the river if you raise because you feel that now you have the best hand?

[Ed Miller]

But what happens on the river if you raise because you feel that now you have the best hand?

I don't understand the question.

[Mikey]

Ed you have to explain this balancing effect in more detail, by merely stating "balancing effect" I don't understand it. I guess I'm just seeing it my way.

Open this door for me.

[Mikey]

Don't you rodds change (effectively) if you were to raise the river, believing you have the best hand?

[Ed Miller]

Point is, sometimes you lose a bet on the river when you "improve" but are still actually behind. Other times, you win an extra bet on the river when you improve, and your opponent pays you off (or bets with the worse hand). These bets tend to balance one another.

[Mikey]

"BTW, I thought this through before I wrote the Hidden Outs section... and I made a conscious decision not to mention this wrinkle at all."

I'm going to look for more wrinkles in the book. I think they should stir up worthy discussion.

[Ed Miller]

Don't you rodds change (effectively) if you were to raise the river, believing you have the best hand?

I guess. But that is totally not relevant to whether calling the turn is right or not. Or at least it could only make you more likely to call the turn, not less.

After all, you'd only raise if it were +EV, wouldn't you?

[Guy McSucker]

[ QUOTE ]

After all, you'd only raise if it were +EV, wouldn't you?


[/ QUOTE ]

Exactly.

Which means that you have the best hand more than half the time when you are called. Since you won't always be called, that means you must think you have the best hand more than half the time, and then some...

Which means that you're only raising when the "out" that hits is one of your cleanest ones, one that will be good much more than half the time. If you have some very clean outs like that, you can be all the more willing to call the turn bet, since you now have some implied odds working in your favour, to further counter the effective odds which hurt you.

Guy.

P.S. Yay, I finally understood something in Ed's book.

[DonkeyKong]

I think one other way to think about it Mikey if you want to do it your way is to take all your odds scenarios and do a weighted average. The point is that the average of all of these is +EV. There is a good deal of variance in the result but that doesn't change your betting much. You are playing this hand as you would play it if you faced it 1000+ times, not just once. If you averaged all the results of all those scenarios, you would make money by calling.

Ed, the hidden outs section goes through this weighted avg scenario thing but doesn't give a short-handed way to quickly calculate this in your head... Should you just add ~2 outs (conservatively) to your existing outs when your kicker is bigger than all the cards on the board and maybe 1 when it isn't???

thanks for your response