SNGPT says it's negEV....

[Sykes]

[ QUOTE ]
[ QUOTE ]
Good analysis AggieYank. This type of thinking is much more valuable than EV estimates spit out by SnGPT, which are too sensitive to assumptions (GIGO) and can't cover all the variables.

To me, this is a "must play" for the reasons AggieYank put forth, but it is not a push, not yet at least. If you push now, BB could fold in the hope you will get knocked out. He may be unlikely to fold, but it also may be that's his only chance if you push and he has two worthless cards.

[/ QUOTE ]

Huh? I think your action reads need a little work. If you push, BB is folding effectively 100% of the time.

eastbay

[/ QUOTE ]

except this time, right?

[gumpzilla]

[ QUOTE ]

Huh? I think your action reads need a little work. If you push, BB is folding effectively 100% of the time.

[/ QUOTE ]

I see this situation extremely rarely, so I don't have a good idea. But I'd think that BB would need at least an average level of savviness to put this together. I think he can fold with a lot of garbage, but I wouldn't be surprised to see many BBs at the 22s calling with 66 or A8 type holdings, either.

[pooh74]

You're looking to double up, and I see this as the perfect opportunity

This is a great quote because I think it does a good job of demonstrating what things like ICM and SNGPT do. You should be able to feel this, but doing the math helps. Your main objective at this stage is not looking for double ups with these stacks.

Sorry if I'm taking your stuff out of context, but this is the part of the game where double ups need to be near certain for them to be worth the risk. Stack disparity, stack disparity.

[eastbay]

[ QUOTE ]
PP $22, I plugged in CO pushing instead of raising, SnGPT says I should be folding QQ to a maniacal push range, and KK to a loose push range....what do you guys think?


PartyPoker, Big Blind is t150 (4 handed) Converter on pregopoker.com

CO (t2670)
Hero (t915)
SB (t4135)
BB (t280)

Preflop: Hero is in Button with Q[img]/images/graemlins/spade.gif[/img] Q[img]/images/graemlins/heart.gif[/img]
<font color="red">CO raises t520</font>, <font color="red"> Hero raises 915 (all in) </font>, <font color="gray">SB folds</font>, <font color="red"> BB calls 130 </font>, CO calls t395

[/ QUOTE ]

There's a lot of stuff going on here, and a lot of the replies don't make much sense to me.

First, depending on how CO has been playing the default "maniac" range of 22+,A2s+,A3o+,KTs+,KJo+,QJs is probably considerably too tight. If big stack is passive, CO might raise about top 50% here, being happy to go in as a slight favorite with BB's random hand to eliminate him.

Second, there's a weakness to the analysis here that falls into the "ultra-short stack" category. If you fold, SnGPT assumes BB will also fold. This isn't a good assumption here, since BB will call a lot of the time, and that has important consequences for this bubble hand.

Since he's more or less out if he folds, and he's probably near 50/50 to double up or bust if he calls, then the analysis isn't considering the 50% chance that he doubles if you fold. This lowers your "Fold EV" quite a bit, since we're on the bubble.

Therefore, I think the analysis is overtightening your call range here due to the aforementioned effects. I call the QQ in practice, although I think I could possibly be convinced otherwise if the analysis was expanded to include the BB's call properly.

Considering action behind your fold is in the plans for improving SnGPT, but it's kind of a complex thing to do it well.

eastbay

[ZeroPointMachine]

[ QUOTE ]
[ QUOTE ]
PP $22, I plugged in CO pushing instead of raising, SnGPT says I should be folding QQ to a maniacal push range, and KK to a loose push range....what do you guys think?


PartyPoker, Big Blind is t150 (4 handed) Converter on pregopoker.com

CO (t2670)
Hero (t915)
SB (t4135)
BB (t280)

Preflop: Hero is in Button with Q[img]/images/graemlins/spade.gif[/img] Q[img]/images/graemlins/heart.gif[/img]
<font color="red">CO raises t520</font>, <font color="red"> Hero raises 915 (all in) </font>, <font color="gray">SB folds</font>, <font color="red"> BB calls 130 </font>, CO calls t395

[/ QUOTE ]

There's a lot of stuff going on here, and a lot of the replies don't make much sense to me.

First, depending on how CO has been playing the default "maniac" range of 22+,A2s+,A3o+,KTs+,KJo+,QJs is probably considerably too tight. If big stack is passive, CO might raise about top 50% here, being happy to go in as a slight favorite with BB's random hand to eliminate him.

Second, there's a weakness to the analysis here that falls into the "ultra-short stack" category. If you fold, SnGPT assumes BB will also fold. This isn't a good assumption here, since BB will call a lot of the time, and that has important consequences for this bubble hand.

Since he's more or less out if he folds, and he's probably near 50/50 to double up or bust if he calls, then the analysis isn't considering the 50% chance that he doubles if you fold. This lowers your "Fold EV" quite a bit, since we're on the bubble.

Therefore, I think the analysis is overtightening your call range here due to the aforementioned effects. I call the QQ.

Considering action behind your fold is in the plans for improving SnGPT, but it's kind of a complex thing to do it well.

eastbay

[/ QUOTE ]

I don't understand how you can expand CO range when he is pushing 17+BB into the big stack. I also think BB has much less than 50% chance of doubling up with his random hand against CO's range.

I do agree that there are many circumstances where SNGPT over states your fold equity and this leads to misleading situations. Unfortunately, most of these situations involve calculating future actions and get complicated very quickly.

Wouldn't the fact that BB could be eliminated a large percentage of the time if you stay out of the hand actually increase your fold equity in this case.

How about just flat calling and folding to a CO bet with an overcard flop?

[eastbay]

[ QUOTE ]

I don't understand how you can expand CO range when he is pushing 17+BB into the big stack.


[/ QUOTE ]

Without any more information I consider the raise to be weakness (I'm better than BB but I want a way to get out if SB picks something up.)

[ QUOTE ]

I also think BB has much less than 50% chance of doubling up with his random hand against CO's range.


[/ QUOTE ]

Well clearly this depends on point 1.

[ QUOTE ]

I do agree that there are many circumstances where SNGPT over states your fold equity and this leads to misleading situations. Unfortunately, most of these situations involve calculating future actions and get complicated very quickly.


[/ QUOTE ]

True.

[ QUOTE ]

Wouldn't the fact that BB could be eliminated a large percentage of the time if you stay out of the hand actually increase your fold equity in this case.

[/ QUOTE ]

Well first of all I don't think it's a large percentage of the time. Second, no, not by much, since if BB folds his 150 chips away, he's darn near busted anyway.

eastbay

[ZeroPointMachine]

This is an auto fold. Yes, you are well ahead of his range, but you have to understand that when you lose here you are losing $11.50 in equity. When you win here you only have $15.00 equity in the tourney. If cutoff takes out the short stack your equity jumps up to $12.65.

Sometimes you are going to make the money as a short stack. This is not a crime. Trying to hard to avoid it is just giving money away.

[eastbay]

[ QUOTE ]
This is an auto fold. Yes, you are well ahead of his range, but you have to understand that when you lose here you are losing $11.50 in equity. When you win here you only have $15.00 equity in the tourney. If cutoff takes out the short stack your equity jumps up to $12.65.

Sometimes you are going to make the money as a short stack. This is not a crime. Trying to hard to avoid it is just giving money away.

[/ QUOTE ]

Care to justify that with something more than an assertion?

Here's my expanded analysis:

I like the $EV call number from SnGPT of 22.6% because BB will almost certainly fold, hoping you will bust. SB isn't going to call without a monster.

If you fold, there are several possibilities. Let's start by assuming that SB is a $22 weakie and will fold.

Then let's assume that BB always calls on pot odds, and is 50/50 in the hand. If he busts, your equity is 25.3%, if he doubles, your equity is 18.6%, the mean of which is 21.9%.

22.6% &gt; 21.9%, so it looks like a call with those assumptions.

If BB calls sometimes and folds, sometimes, then the fold equity is a blend of 21.9% and the SnGPT number of 23.0%. This implies that the breakeven point is BB folding 63% of the time. That seems high, although this is a $22 so anything is possible. In any case, it seems a lot closer than "auto fold" to me. How do you arrive at "auto fold"?

eastbay

[ZeroPointMachine]

[ QUOTE ]
[ QUOTE ]
This is an auto fold. Yes, you are well ahead of his range, but you have to understand that when you lose here you are losing $11.50 in equity. When you win here you only have $15.00 equity in the tourney. If cutoff takes out the short stack your equity jumps up to $12.65.

Sometimes you are going to make the money as a short stack. This is not a crime. Trying to hard to avoid it is just giving money away.

[/ QUOTE ]

Care to justify that with something more than an assertion?

Here's my expanded analysis:

I like the $EV call number from SnGPT of 22.6% because BB will almost certainly fold, hoping you will bust. SB isn't going to call without a monster.

If you fold, there are several possibilities. Let's start by assuming that SB is a $22 weakie and will fold.

Then let's assume that BB always calls on pot odds, and is 50/50 in the hand. If he busts, your equity is 25.3%, if he doubles, your equity is 18.6%, the mean of which is 21.9%.

22.6% &gt; 21.9%, so it looks like a call with those assumptions.

If BB calls sometimes and folds, sometimes, then the fold equity is a blend of 21.9% and the SnGPT number of 23.0%. This implies that the breakeven point is BB folding 63% of the time. That seems high, although this is a $22 so anything is possible. In any case, it seems a lot closer than "auto fold" to me. How do you arrive at "auto fold"?

eastbay

[/ QUOTE ]

Well, first of all I'm half asleep and misread the HH. I thought CO pushed. [img]/images/graemlins/blush.gif[/img] [img]/images/graemlins/blush.gif[/img]

However, I still think your range for CO is too large given the big stack in the SB. CO has a hand that he likes against the small stack and I doubt it is T8s. Estimating how often the BB calls is of course a guess at best in a 22.