Small Stakes Shorthanded NL Poker Forums

Page 3: nl100 3 way 3 bet pot OOP

or track by Email or RSS


Sounded Simple

Avatar for Sounded Simple

1009 posts
Joined 03/2008

If I ever come across somebody who's actually balanced, I'll let you know Wink[/b]

Andrew



This is flawed thinking.

Sometimes:
(+EV Turn Call) + (+EV River Call) = -EV

At the heart of the misconception is that on the river that the villain to be balanced, must bet some precise amount - lets say 60%
In a vacuum once we arrive at the every river there will be this point based on pot odds, at 59% bluff frequency we should fold at 61% pot odds we should call.

However when we calculate EV across two streets it's much easier for the villain to have a bluff frequency on the river that puts us in a shitty spot.
For example the villain might be "balanced" (i.e. hero loses money regardless whether he calls or folds) if villains bluff frequency is anywhere between 30%-90%.

So even fairly poor non thinking players, even fish can quite often find balance against you - even by mistake.


This might sound like we are getting on your case here but I think, trust me that it's not true. I've learned much from your material and really respect your approach to the game both as a player and a coach.
Also, it's not that you are alone in this misconception most players think similarly but since you think deeply write a lot about theory I think it's worth discussing Smile

I know it's a mindfuck but you can trust me (and Stheif09) on the math side of things.
There's tonnes of stuff in this concept to think about, it's difficult to internalise but I'll gladly help if you want clarification on anything or want to debate that the concept is correct.

Posted over 2 years ago

Sounded Simple

Avatar for Sounded Simple

1009 posts
Joined 03/2008

Also, sorry to OP for the huge sidetrack.

Fwiw, I agree with those who like to make it smaller 3b and cbet.
Turn I think is c/f since without a read he shouldn,t have very many pure floats here and I doubt you ever get a fold from pair with the stack/pot ratio as it is.

If you had made preflop and flop smaller then there may be merit in firing 3 barrels, but that's another question.

Posted over 2 years ago

BalugaWhale

Avatar for BalugaWhale

1087 posts
Joined 01/2008

This is flawed thinking.

Sometimes:
(+EV Turn Call) + (+EV River Call) = -EV

At the heart of the misconception is that on the river that the villain to be balanced, must bet some precise amount - lets say 60%
In a vacuum once we arrive at the every river there will be this point based on pot odds, at 59% bluff frequency we should fold at 61% pot odds we should call.

However when we calculate EV across two streets it's much easier for the villain to have a bluff frequency on the river that puts us in a shitty spot.
For example the villain might be "balanced" (i.e. hero loses money regardless whether he calls or folds) if villains bluff frequency is anywhere between 30%-90%.

So even fairly poor non thinking players, even fish can quite often find balance against you - even by mistake.


This might sound like we are getting on your case here but I think, trust me that it's not true. I've learned much from your material and really respect your approach to the game both as a player and a coach.
Also, it's not that you are alone in this misconception most players think similarly but since you think deeply write a lot about theory I think it's worth discussing Smile

I know it's a mindfuck but you can trust me (and Stheif09) on the math side of things.
There's tonnes of stuff in this concept to think about, it's difficult to internalise but I'll gladly help if you want clarification on anything or want to debate that the concept is correct.



I'll gladly defer to you guys on the math.

Can +EV preflop call--> +EV flop call --> +EV turn call --> +EV river call = -EV?

Andrew

Posted over 2 years ago

Swaytje

Avatar for Swaytje

212 posts
Joined 07/2008

@ Andrew

Yes it can, read this post by AGtJ Smile

Here's another simple illustration I found on 2p2 a while back that shows that you can call every street (including the river) with "correct" immediate odds, and it's a very -EV play.

"Let's say we have like A8s on the BTN and call a raise from the CO. We know he pots all flops turns and rivers, but we also know that we always (on all flops, turns and rivers) have 35% equity aginst his range. We're also 100 bb's deep.

So, when he pots flop, we clearly call. Since he pots, we only need 33% equity to continue. So it's +EV to call.

Turn, we obviously also call. Since he pots, we still only need to win 33% of the time, and we have 35% equity. So it's +EV to call.

On the river, we call his pot sized shove. Once again, we only need 33% equity and we still have 35%. So clearly, calling river is also gonna be +EV.

So we've made 3 +EV decisions and we're happy with ourselves. We decide to sit down and actually calculate how much we're winning long term by playing this good.

Pre, he bets 3 bb's. Flop he bets 7.5bb's. Turn he bets 22.5 bb's. Since we only 100bb's deep, he bets his remaining 67bb's. That's a total of 100bb. We also win 1.5 bb's from the blinds when we win.

So, 35% of the time we win 101.5bb's. EV = 0.35*101.5 = +35.525
And 65% of the time we lose 100bb's. EV = 0.65*100 = -65

Total EV of calling down = 35.525 - 65 = -29.475"

I'm out til later, maybe sounded simple can pick up for me.

Posted over 2 years ago

marco

Avatar for marco

730 posts
Joined 08/2010


"Let's say we have like A8s on the BTN and call a raise from the CO. We know he pots all flops turns and rivers, but we also know that we always (on all flops, turns and rivers) have 35% equity aginst his range. We're also 100 bb's deep.

.....

Pre, he bets 3 bb's. Flop he bets 7.5bb's. Turn he bets 22.5 bb's. Since we only 100bb's deep, he bets his remaining 67bb's. That's a total of 100bb. We also win 1.5 bb's from the blinds when we win.

So, 35% of the time we win 101.5bb's. EV = 0.35*101.5 = +35.525
And 65% of the time we lose 100bb's. EV = 0.65*100 = -65

Total EV of calling down = 35.525 - 65 = -29.475"




The math shown does not describe the situation as stated. 35% equity on all streets does not equate to a 35% chance of winning.

You have to add up the EV of each individual spot. Edit: or that's what i thought anyway. Regardless, the math above is not right. I'm trying to model what you are talking about, w/o luck.

Posted over 2 years ago

Ass Get to Jigglin

Avatar for Ass Get to Jigglin

4276 posts
Joined 10/2010

The math shown does not describe the situation as stated. 35% equity on all streets does not equate to a 35% chance of winning.

You have to add up the EV of each individual spot. Edit: or that's what i thought anyway. Regardless, the math above is not right. I'm trying to model what you are talking about, w/o luck.



pretty sure that it's what Andrew was asking when he said "+EV flop call" etc across all streets. I think he was talking about your equity vs his range compared to the *immediate* odds. I also think this is what Sounded Simple meant by, "Sometimes: (+EV Turn Call) + (+EV River Call) = -EV " If I'm wrong, I'd ask him to please correct me.



The sauce post I copied and pasted is a really good read where he gives a decent approach to roughly modeling a turn + river situation.

Posted over 2 years ago

Sounded Simple

Avatar for Sounded Simple

1009 posts
Joined 03/2008

I'll gladly defer to you guys on the math.

Can +EV preflop call--> +EV flop call --> +EV turn call --> +EV river call = -EV?

Andrew



Yes, as Swaytje's requoted post from AGtJ shows.


It gets worse too...
Sometimes you can have 50%+ on the earlier street and the concept still applies
:|

Posted over 2 years ago

Sounded Simple

Avatar for Sounded Simple

1009 posts
Joined 03/2008

The math shown does not describe the situation as stated. 35% equity on all streets does not equate to a 35% chance of winning.

You have to add up the EV of each individual spot. Edit: or that's what i thought anyway. Regardless, the math above is not right. I'm trying to model what you are talking about, w/o luck.




You have actually kind of reached the right conclusion, AGtJ is correctly pointing out the flaw that is in conventional thinking in the post you refer to.

On early streets pure "pokerstove" equity is really only applicable when you are all-in or likely to be all-in on that street.

Having 10%, 25% or 55% equity on the flop isn't the be all and end all, it's massively important of course but what really matters is the global EV of all play on all streets combined.

That doesn't have to mean super complicated math, it just means you have to look a couple of streets ahead and factor in who is going to have the best of it there.

As an example, if you had an opponent you knew would vbet light but give you credit for a river raise then you might have a very high EV for river play so you would be happy to make a thin turn call.

Posted over 2 years ago

BalugaWhale

Avatar for BalugaWhale

1087 posts
Joined 01/2008

You have actually kind of reached the right conclusion, AGtJ is correctly pointing out the flaw that is in conventional thinking in the post you refer to.

On early streets pure "pokerstove" equity is really only applicable when you are all-in or likely to be all-in on that street.

Having 10%, 25% or 55% equity on the flop isn't the be all and end all, it's massively important of course but what really matters is the global EV of all play on all streets combined.

That doesn't have to mean super complicated math, it just means you have to look a couple of streets ahead and factor in who is going to have the best of it there.

As an example, if you had an opponent you knew would vbet light but give you credit for a river raise then you might have a very high EV for river play so you would be happy to make a thin turn call.



So I'm 100% confident you guys are right about all of this, and I think intuitively a lot of us get there without the math (though the math certainly explains and helps!)

However, there is most certainly a point at which our opponent is bluffing *too* much on a street that no matter how balanced they are on the later street, we shouldn't fold.

For example, if our opponent has 100 combos on the turn, and 99 of them are drawing dead, and 1 of them has us drawing dead, we can call without fear of reverse implied odds from the one remaining (there should be no combination of bluffs/values on the river that make our turn call bad in this case, right?)

I'm sure this discussion is helping many other people along with myself Smile

Andrew

Posted over 2 years ago

Sounded Simple

Avatar for Sounded Simple

1009 posts
Joined 03/2008

Yes if we have him 99% of the time then even if he achieves the most perfect balance he still has to give up the pot something like 98% there.

Posted over 2 years ago

BalugaWhale

Avatar for BalugaWhale

1087 posts
Joined 01/2008

Yes if we have him 99% of the time then even if he achieves the most perfect balance he still has to give up the pot something like 98% there.



alright cool, so what that means is:

1) He can still be bluffing so much on the turn that we should not fold *despite* the presence of reverse implied odds on the river

2) the amount he needs to be bluffing starts high and gets progressively lower from street to street.

3) he needs to be bluffing a fair amount more than the direct odds would indicate

very cool

Andrew

Posted over 2 years ago

jamesk777

Avatar for jamesk777

5 posts
Joined 11/2010

thanks for the input guys. I learned alot!

Posted over 2 years ago

nawhead

Avatar for nawhead

2485 posts
Joined 10/2009

The math shown does not describe the situation as stated. 35% equity on all streets does not equate to a 35% chance of winning.

You have to add up the EV of each individual spot. Edit: or that's what i thought anyway. Regardless, the math above is not right. I'm trying to model what you are talking about, w/o luck.


yeah, i agree.

this:
EV = 0.35*101.5 = +35.525
EV = 0.65*100 = -65

is wrong since EV calcs have to use the whole pot, and any bets put in by us on a previous street don't get disappeared/ignored like this.

if the above was correct, then we'd be doing EV calcs something like this:

EV flop = .35win (7.5+ 7.5 - 3) - .65loss (7.5) = 4.2 - 4.65 = -0.45
EV turn = .35 (22.5 + 22.5 - 7.5 - 3) - .65 (22.5) = 12.075 - 14.625 = -2.55
EV rivr = .35 (67.5 + 67.5 - 22.5 - 7.5 - 3) - .65 (67.5) = 35.7 - 43.875 = -8.175

EV hand = -0.45 + -2.55 + -8.175 = -11.175

instead of:

EV flop = .35win (7.5+7.5) - .65loss (7.5) = 5.25 - 4.65 = 0.6
EV turn = .35 (22.5 + 22.5) - .65 (22.5) = 15.75 - 14.625 = 1.125
EV rivr = .35 (67.5 + 67.5) - .65 (67.5) = 47.25 - 43.875 = 3.375

EV hand = 0.6 + 1.125 + 3.375 = 5.1

Posted over 2 years ago

Ass Get to Jigglin

Avatar for Ass Get to Jigglin

4276 posts
Joined 10/2010

the EV of calling down 3 streets in the above scenario of having 35% equity on each street is definitely not +EV, it's hugely -EV.

hint - you have to factor in turn and river action when you calculate flop EV, or if your just on the turn you have to factor in river action when you are calculating turn EV, not just calculate all-in EV on each individual street and add them up.

Posted over 2 years ago

Swaytje

Avatar for Swaytje

212 posts
Joined 07/2008

AGtJ is right, although it might seem very weird. Mostly because we've all learned the normal way of calculating EV (which is good, but doesn't consider reveversed implied odds at all).

Just think of it like this....

You and the villain (assuming its a HU pot) are both putting in half the money each.. Lets say 100$ each.
You're putting all your money in with 35% equity (in this example that is).
How can you turn a profit by putting in 50% of the total pot while only having 35% equity?
Sure, the decisions in a vacuum are +EV. But this is really only accurate on streets prior to the river when going allin (hence no more bets on future streets).

Posted over 2 years ago




HomePoker ForumsSmall Stakes Shorthanded NL → nl100 3 way 3 bet pot OOP