I would fold and steal on the next hand

[Phil Van Sexton]

On several posts recently where calling is slightly +$EV, I see a lot of responses like "I'd fold and steal on the next hand." I decided to look into this plan.....

Blinds 100/200. Party 30/3

UTG 2000
BUTTON 1700
SB 2100
BB/HERO 2200

UTG folds, button pushes, SB folds.

You have XX. You run ICM and find that calling gives and ICM of 0.26 and
folding is 0.25.

This is close, so you decide to just fold and push on the next hand from the SB.
You know the player to your left is tight and will only call with AA-77, AK, and
AQ.

A random hand against this range is 29.4%. However, there is only a 8.28%
chance that he'll be dealt one of these (ie 14/169).

Stacks are now...
UTG 2000
BUTTON 2000
SB/HERO 2000
BB 2000

If you push and he folds, your ICM is 0.2657.
If you push and he calls and you win, your ICM is 0.3833. (It's 0 if you lose,
of course).

(0.2657 * 0.917)+((0.3833*0.294)*0.0828) = 0.253

On the previous hand we turned turn an ICM of 0.26 so we could get an ICM of 0.253.

Well, what if UTG or button push? If they push and we fold, our ICM is 0.2425.

If they push 60% of the time and we always fold, your ICM is:
(0.2425 * 0.40) + (0.253 * 0.60) = 0.2467

Of course, sometimes you'll get a monster hand and call their steal, or the BB will call the steal. This is too complicated. If you throw in a few decimals for this, you get back to 0.25. Remember, we were 0.253 if the UTG/Button never pushed.

Let's review. You were in BB. You could have called for 0.26 ICM. Instead, you folded for 0.25 with the intention of pushing on the next hand to make up for this. You plan has an ICM of 0.25, an increase of 0.

Isn't blind stealing good? Why didn't the ICM go up? Well, a blind steal at 100/200 only has a small influence on ICM. You are risking your whole stack to win only 200 after all. It is+$EV, but not enough to make up for the fact that you have to post the SB.

Moral of the story: blinds steals give small increases in $EV. When people say, "I'll pass on the small +$EV because I'd rather blind steal", they are contradicting themselves.

[Freudian]

[ QUOTE ]

Moral of the story: blinds steals give small increases in $EV. When people say, "I'll pass on the small +$EV because I'd rather blind steal", they are contradicting themselves.

[/ QUOTE ]

By itself, it is of course true. But the accumulated effect of a sound blind stealing strategy is a large increase in +EV.

Thats not to say calling a small +EV situation is wrong on the bubble of course.

[1C5]

That is at 100/200 blinds. What about 150 and 300 blinds or 200/400. That must change things quite a bit?

[Scuba Chuck]

I like what you're trying to prove. I think you should change two things.

1) Probability that the hand will be folded to you, so you can attempt a steal

2) Probability BB has a hand AA-77, AK is actually 2.5% (not 8%) because there is only 6 ways to make pairs.

Reducing the probability of BB having this hand, and the probability it's folded to you is probably equal to or more -$EV than what you're selling here. So it should still prove your point (if not help it).

[ilya]

[ QUOTE ]

2) Probability BB has a hand AA-77, AK is actually 2.5% (not 8%) because there is only 6 ways to make pairs.

Reducing the probability of BB having this hand, and the probability it's folded to you is probably equal to or more -$EV than what you're selling here. So it should still prove your point (if not help it).

[/ QUOTE ]

Umm, what? Reducing the probability of BB holding a big hand from 8% to 2.5% definitely increases your $EV for that push, and thus HURTS his point.

Also, I dunno how you guys got your numbers, but 2.5% and 8% are both wrong. It's about 6% (a little less because sometimes you'll be holding one of his cards & he'll have less than the full 80 ways to make his hand).

[Scuba Chuck]

[ QUOTE ]
Umm, what? Reducing the probability of BB holding a big hand from 8.5% to 2.5% definitely increases your $EV for that push, and thus HURTS his point.

[/ QUOTE ] .

Incorrect. What is the probability it's folded to hero? Do you think there's a 94% probability it's folded to hero. I seriously doubt it.

[ilya]

[ QUOTE ]
[ QUOTE ]
Umm, what? Reducing the probability of BB holding a big hand from 8.5% to 2.5% definitely increases your $EV for that push, and thus HURTS his point.

[/ QUOTE ] .

Incorrect. What is the probability it's folded to hero? Do you think there's a 94% probability it's folded to hero. I seriously doubt it.

[/ QUOTE ]

My point is correct. Your $EV with a random hand is higher when BB's call% is lower. The fact that you may not always get a chance to push doesn't change that.

Edit: ok, I see what you're saying in your OP, but I disagree. I do think UTG & button will both fold something like 40% of the time. Actually I think they will both fold something like 50% of the time.

[The Yugoslavian]

[ QUOTE ]
[ QUOTE ]
Umm, what? Reducing the probability of BB holding a big hand from 8.5% to 2.5% definitely increases your $EV for that push, and thus HURTS his point.

[/ QUOTE ] .

Incorrect. What is the probability it's folded to hero? Do you think there's a 94% probability it's folded to hero. I seriously doubt it.

[/ QUOTE ]

Whatchu talkin' bout Scuba dooba doo?! I'm not sure any of Ilya's point has anything to do with how much it is or isn't folded to hero...

Yugoslav
Who likes the idea of factoring in how often it will be folded to Hero....

[Scuba Chuck]

[ QUOTE ]
Edit: ok, I see what you're saying in your OP, but I disagree. I do think UTG & button will both fold something like 40% of the time. Actually I think they will both fold something like 50% of the time.


[/ QUOTE ]

I'm pretty confident you're making Phil VS's point even stronger. If you think there's only a 50% chance of it getting folded you on the next hand, then the advice "I'd fold and steal on the next hand." is beginning to look even worse. Or in this case -$EV.

FWIW, I was suggesting that I don't think there is a 94% probability that the next hand will be folded to hero. Which it looks like we completely agree on.

[ilya]

Yes, but Phil's rough calculations assumed that it would only be folded to you 40% of the time. Plus he had that wacky 8.28% figure.

[Scuba Chuck]

[ QUOTE ]
Yes, but Phil's rough calculations assumed that it would only be folded to you 40% of the time. Plus he had that wacky 8.28% figure.

[/ QUOTE ]

I can't find that 40% number. I think he's assuming it will be folded to you 100% of the time.

Regardless, I think he has a very valid point.

Ilya, are you arguing against his point? If so, why?

[pbutkus]

i think the main prob w this is the calling standards. at the 30s on pp aa-77, ak, aq only is not true. you'll 100% get called by any pair and any ace, and most likely any two facecards, as well as most kings. how does that effect percentages.

[ilya]

[ QUOTE ]
[ QUOTE ]
Yes, but Phil's rough calculations assumed that it would only be folded to you 40% of the time. Plus he had that wacky 8.28% figure.

[/ QUOTE ]

I can't find that 40% number. I think he's assuming it will be folded to you 100% of the time.

Regardless, I think he has a very valid point.

Ilya, are you arguing against his point? If so, why?

[/ QUOTE ]

The 40% part is in this bit:

[ QUOTE ]

If you push and he folds, your ICM is 0.2657.
If you push and he calls and you win, your ICM is 0.3833. (It's 0 if you lose,
of course).

(0.2657 * 0.917)+((0.3833*0.294)*0.0828) = 0.253

On the previous hand we turned turn an ICM of 0.26 so we could get an ICM of 0.253.

Well, what if UTG or button push? If they push and we fold, our ICM is 0.2425.

If they push 60% of the time and we always fold, your ICM is:
(0.2425 * 0.40) + (0.253 * 0.60) = 0.2467

Of course, sometimes you'll get a monster hand and call their steal, or the BB will call the steal. This is too complicated. If you throw in a few decimals for this, you get back to 0.25. Remember, we were 0.253 if the UTG/Button never pushed.


[/ QUOTE ]

I'm not exactly sure what Phil's point is, so I wouldn't say I'm arguing against it.

[microbet]

dadadadada
communication breakdown,
It’s always the same,
I’m having a nervous breakdown,
Drive me insane!

[The Yugoslavian]

[ QUOTE ]
dadadadada
communication breakdown,
It’s always the same,
I’m having a nervous breakdown,
Drive me insane!

[/ QUOTE ]

Finally...someone talking sense in this thread!!

Phil started us out right...but then things went horribly, horribly wrong!!

Yugoslav

[ilya]

[ QUOTE ]
[ QUOTE ]
dadadadada
communication breakdown,
It’s always the same,
I’m having a nervous breakdown,
Drive me insane!

[/ QUOTE ]

Finally...someone talking sense in this thread!!

Phil started us out right...but then things went horribly, horribly wrong!!

Yugoslav

[/ QUOTE ]

What are you saying here Yugo...I don't understand.... /images/graemlins/confused.gif

[Phil Van Sexton]

[ QUOTE ]
Phil started us out right...but then things went horribly, horribly wrong!!

[/ QUOTE ]

I knew I would screw up some of those numbers, but that's never stopped me before.

The BB calling % of 8.28% was clearly wrong. I think its 5.55%. If someone has a better way to calculate this, clue me in. I went here (http://www.poker1.com/mcu/tables/Table18.asp) and found a specific pair is 0.45%, a suited hand is 0.3%, and unsuited is 0.90%. We have 7 pairs and 2 suited and unsuited (AK, AQ).

Anyway, here is the corrected ICM of the blind steal assuming we always fold if the Button or UTG try to steal. I assumed they would steal 60% of the time. That number is pretty arbitrary, so here are the result with several numbers:

Button/UTG steal 0% = 0.257 (was 0.253)
Button/UTG steal 20% = 0.254
Button/UTG steal 40% = 0.251
Button/UTG steal 60% = 0.248 (was 0.2467)

Slightly better, but not a huge difference.

[Scuba Chuck]

Phil here's some numbers for you:

There is 6 ways to make any pocket pair
There are 13 Total pocket pairs

Pairs = 13*6 = 78 of total possible hand combinations

There are 16 possible ways of making any other two card combo.
2 card combos = (169-13)*16 = 2,496

Total possible card combinations = 2,496+78= 2,574

Therefore AA-77 = 48 possible card combinations
AK = 16
Total = 64

Probability of getting this = 64/2574 = 2.49%

[Scuba Chuck]

[ QUOTE ]
how does that effect percentages.

[/ QUOTE ]

It proves Phil's point even more. Any two cards vs. those hands is going to decrease $EV.

I think the overall belief here is failing in two methods.

1) Belief that skill can overcome alternate +$EV situations. I hashed this out a little while back. See What to do when ICM fails you? (http://forumserver.twoplustwo.com/showthreaded.php?Cat=&Board=singletable&Number=202 1562&Forum=f22&Words=ICM&Searchpage=0&Limit=25&Mai n=2021562&Search=true&where=sub&Name=22022&dateran ge=1&newerval=3&newertype=w&olderval=&oldertype=&b odyprev=#Post2021562)

2) Overvaluing Survival Equity

Scuba

[Paul2432]

[ QUOTE ]
Phil here's some numbers for you:

There is 6 ways to make any pocket pair
There are 13 Total pocket pairs

Pairs = 13*6 = 78 of total possible hand combinations

There are 16 possible ways of making any other two card combo.
2 card combos = (169-13)*16 = 2,496

Total possible card combinations = 2,496+78= 2,574

Therefore AA-77 = 48 possible card combinations
AK = 16
Total = 64

Probability of getting this = 64/2574 = 2.49%

[/ QUOTE ]

There are not 16 ways of making any other two card combo. There are 12 ways of making an offsuit hand and 4 ways of making a suited hand.

The total number of starting hands is 1326.

Paul

[RobGW]

Phil,
I love the overall jist of this post. I also see too many people passing up +EV in the name of blind stealing. Sure there are situations that this could be correct. But one must consider two things. How tight the two people to your left are. The tighter they are the better to steal. And how aggressive are the players to your right. The more aggressive they are, the less chance you have to steal. If you have aggressive players to your right and loose players to your left you may want to fold and get out of the way. However, my default play is usually to call if its +EV. Once in a while I will deviate from that but I have to have a specific reason to do that. I think too many people overvalue theire ability to steal and overvalue their survival equity. This leads to too much folding and too many 3rds and not enough 1st place finishes.

[Phil Van Sexton]

[ QUOTE ]
[ QUOTE ]
Phil here's some numbers for you:

There is 6 ways to make any pocket pair
There are 13 Total pocket pairs

Pairs = 13*6 = 78 of total possible hand combinations

There are 16 possible ways of making any other two card combo.
2 card combos = (169-13)*16 = 2,496

Total possible card combinations = 2,496+78= 2,574

Therefore AA-77 = 48 possible card combinations
AK = 16
Total = 64

Probability of getting this = 64/2574 = 2.49%

[/ QUOTE ]

There are not 16 ways of making any other two card combo. There are 12 ways of making an offsuit hand and 4 ways of making a suited hand.

The total number of starting hands is 1326.

Paul

[/ QUOTE ]

Paul is right. There are 6 ways to make a pair of Aces:
AcAd, AcAs, AcAh, AdAs, AdAh, AsAh

1326/6 = 221

I'm sure you recognize 220:1 odds for getting aces from various books and other places.

Also, 6/1326 = 0.0045

Just like the 0.45% that I used (the second time).

4 ways to make a suited hand (0.30%), 12 ways for unsuited (0.90%).

We have 7 pairs (AA-77), 2 suited (AKs, AQs) and 2 unsuited (AKo, AQo).

((7 * 6) + (2 * 4) + (2 * 12)) / 1326 = 0.0558

I used 5.55% (different because I rounded some stuff).

The correct answer is 5.58%. Thanks Paul.

[Scuba Chuck]

[ QUOTE ]
I think too many people overvalue theire ability to steal and overvalue their survival equity.

[/ QUOTE ]


Huh, someone actually agrees with me. I'm pretty sure those were almost my words.

[Scuba Chuck]

[ QUOTE ]
There are not 16 ways of making any other two card combo. There are 12 ways of making an offsuit hand and 4 ways of making a suited hand.


[/ QUOTE ]

I'm pretty sure that you just said the same thing. 16 ways to make AK. 12 offsuit, 4 suited.

[ QUOTE ]
The total number of starting hands is 1326.


[/ QUOTE ]

correct. My mistake, I doubled counted the suited vs. offsuited hands.

169 total hand combos.
13 pairs
156 two card combos
Divide that by two and get 78

78*16= 1,248
13*6=78
1,248+78= 1,326

My mistake

So AA-77 = 48
AK = 16

64/1326 = 4.83%

Include AQ = 6.0%

[syka16]

[ QUOTE ]
[ QUOTE ]
I think too many people overvalue theire ability to steal and overvalue their survival equity.

[/ QUOTE ]


Huh, someone actually agrees with me. I'm pretty sure those were almost my words.

[/ QUOTE ]


When I'm around T600 at 50/25 with 7+ players left and know that I'm not going to have a chance to steal due to limping or loose callers to the left, I'm looking for a coinflip with limpers already in and I'm pushing ATo, A9s, KQo, KJs and 77+. I'm not really sure what else to do... Anyone else follow this?

[Scuba Chuck]

syka, I think you're feeling desperate, and there isn't a need to do that until you're sub 5BB.

Furthermore, pushing looks less manacing than a 3-3.5BB raise, IMO