big stack bullying

[housenuts]

when it's on the bubble and you are clearly the big stack at the table how do you effectively bully? does the majority of your steals come from PF raises, or do you like to limp in and steal the pot when you sense weakness on the flop?

i've been stealing pots quite easily with 3x BB raises. whenever i limp it, good players seem to sense weakness from me and raise me right there. this is something i could trap them with if i'm well aware of it.

any suggestions to best weaken their stack for when it's down to 3 or 2 are encouraged.

[SmileyEH]

If you are a big stack and the small stack (if there is one) has folded, pushing with any two is profitable. Basically, if you can put your opponents calling hands down to like AA-JJ + AK then you can go ahead and push with any two every time.

-SmileyEH

[JaBlue]

I raise preflop and if they have the guts to tackle with me generally raise the flop with any two too. I'm just a bastard, though /images/graemlins/smile.gif.

note: use sparingly. they will catch on after 3 or 4 times. Then play better hands.

[housenuts]

[ QUOTE ]
If you are a big stack and the small stack (if there is one) has folded, pushing with any two is profitable. Basically, if you can put your opponents calling hands down to like AA-JJ + AK then you can go ahead and push with any two every time.

-SmileyEH

[/ QUOTE ]

i don't know if i like that strategy. you can quickly come from penthouse to outhouse as norm chad would say. i mean you could go from a solid 1st to 3rd.

[therock]

If there are 4 left, push when the 2nd and 3rd place stacks are in the BB. No sense pushing the 4th place stacks BB pre-flop, because he is the most likely to call. I would limp vs. the 4th place player PF then push after the flop.

[Texas Pete]

[ QUOTE ]
If you are a big stack and the small stack (if there is one) has folded, pushing with any two is profitable. Basically, if you can put your opponents calling hands down to like AA-JJ + AK then you can go ahead and push with any two every time.

-SmileyEH

[/ QUOTE ]

I like the idea... can you tell us the math behind it?

[SmileyEH]

Sure...if your opponents are playing that tight they are calling with less than five percent of their hands. This means if you are pushing from the button they are folding nine times out of ten. Now if the blinds are greater than ten percent of their stack you just made a profit - even if you lost every hand that was called (obviously not true).

Example: You (button) 3500 chips, UTG 500 chips, BB and SB each with 2000 chips and the BB is 300.

UTG folds and you push. 19 times the SB folds so you win 9 x 150 = 1650 and you lose 2000 once (assuming you lose every hand. 19 times the BB folds so you win 9 x 300 = 2700 and you lose 2000 once. So you "net profit" is 350 at the very worst. It will actually be much higher because a random hand will win 2
5% of hands against AA-JJ + AK. Therefore your actual profit given these numbers is 350 + .25*2*2000 = ~1350.

-SmileyEH

[tigerite]

.. and some really tight players may fold JJ.
Don't laugh, it happened to me, THREE handed once, when I pushed from UTG with Q7s, and the SB (2nd biggest stack) called with A5o or some rubbish. A jack came, as did an A, but I made a straight. LOL.

[housenuts]

what happens if they aren't so tight?

[UMTerp]

[ QUOTE ]
what happens if they aren't so tight?


[/ QUOTE ]

Then you sit back while they bust each other out and play one of them heads up at the end.

[SmileyEH]

This is horrible strategy. Do you see why?

-SmileyEH

[tigerite]

If they are loose, you tighten up a bit, but not ridiculously so. You push with about the top 30% of hands instead of the top 100%. At least, that's what I'd do.

[ilya]

[ QUOTE ]

UTG folds and you push. 19 times the SB folds so you win 9 x 150 = 1650 and you lose 2000 once (assuming you lose every hand. 19 times the BB folds so you win 9 x 300 = 2700 and you lose 2000 once. So you "net profit" is 350 at the very worst. It will actually be much higher because a random hand will win 2
5% of hands against AA-JJ + AK. Therefore your actual profit given these numbers is 350 + .25*2*2000 = ~1350.

-SmileyEH

[/ QUOTE ]

I think your math may be off Smiley.
If your opponents call with 5% of their hands each (which, incidentally, is not AA-JJ/AK but AA-TT/AK-AQ), that means you get called about 10% of the time. So far so good, same thing you said.
But the profit, let's say for 32o: (don't forget, you are never actually pushing a "random hand")

(0.9)(450) +
(0.1)(0.24)((2300*0.5)+(2150*0.5)) +
(0.1)(0.76)(-2000) =
405 + 53 - 152 =
+306

This is about the worst you can do if both the SB and the BB only call with the top 5% of their hands (just a touch worse than ATo, which is 26.3% against the top 5%).

I'm not sure how you got your 1350 number, but the maximum you could possibly average in that situation is about 565 (you magically have AA every time you push):

(0.9)(450)+
(0.1)(0.85)((2300*0.5)+(2150*0.5))+
(0.1)(0.15)(-2000)=
405 + 190 - 30 =
+565

In response to someone's question about "what if your opponents are looser...", just grab PokerStove and start plugging numbers in. It's a lot of busywork but conceptually quite simple. Pick a range for your opponents, add all those possibilities up, divide by 1326 to see what percentage that range represents. Let's say it's P. Take ((100-P)/100) and multiply it by itself once for each extra opponent after the first. Call that new number C; the percentage chance that one of your opponents will call is simply (100-(100*C)). Etc etc.

[UMTerp]

Agreed. You're still the big stack. I wasn't advocating folding monsters or anything.

[morgan180]

if i did my math correctly even if they loosen up to the top 20% of hands the play is still +ev.

Assuming that when you are called you lose 65% of the time (based on the hands below) and blinds 150/300 now 8 times out of 10 you pick up 450 uncontested and 2 out of 10 you are called and 65% of those two times you lose 2000.

(8*450) - (1.3 * 2000) = 1000

or have i over simplified?

At 20% you are looking at hand ranges like:
AA-66, AKs-A9s, KQs-K9s, QJs-Q9s, JTs-J9s, T9s, AKo-ATo, KQo-KTo, QJo-QTo, JTo

(which now that i look at it is probably unrealistic that they would be willing to call this much... is it really that easy as the big stack late????)

[tigerite]

I gave SB and BB the range:

55+,A5s+,A7+,KTs+,KJ+,QJs,QTs,JTs

which seems reasonable to me. = 17.1% of hands. I made the big stack's hand truly 'random' by using poker calculator.

Wins: 23.8%
Called by either SB or BB: 34.2%


(0.658) * (450) +
(0.342) * (0.238) * ((2300 * 0.5) + (2150 * 0.5)) +
(0.342) * (0.762) * (-2000) =
296.1 + 181.1061 - 521.208 =
-44.0019

As you can see, this is actually -EV.

[morgan180]

thanks - i always sucked at math - but now i've got an excuse to get better at it /images/graemlins/grin.gif

[tigerite]

I think the break-even point for EV is about the top 45% of hands, in this situation. It is certainly +EV if you use the 8 Sklansky-Malmuth groups ( = 40.8% of hands), but not by much, ~20 overall. This does include 32s, but not 32o /images/graemlins/wink.gif

[tigerite]

I've made a mistake somewhere. I'm sure eastbay or ilya will correct it though.. right now I'm too tired.

[morgan180]

could you provide some clarification on the equation please??:

(0.658) <font color="blue">{win uncontested %}</font> * (450)<font color="blue">{amount won}</font> +
(0.342) <font color="blue">{% called}</font>* (0.238) <font color="blue">{% win if called}</font>* ((2300 * 0.5) + (2150 * 0.5)) <font color="blue">{can you explain this part of the equation} </font>+
(0.342) <font color="blue">{% called}</font> * (0.762) <font color="blue">{% you lose}</font> * (-2000) <font color="blue">{amount lost}</font> =
296.1 + 181.1061 - 521.208 =
-44.0019

[ilya]

Thanks for doing the calculations.
However, I think the average SB/BB plays at least a bit tighter. Do you really think many players will call all-in with A9s-A5s/A9o-A7o when they are not particularly short-stacked?
The other interesting thing about this situation is that, if you are confident about a range for your opponents, you need to make sure you're adapting your own range to reflect your read. So, against your range, be happier to push QJo than A6o. But against a modified range that excludes A8s-A5s and A9o-A7o, choose A6o over QJo.

[tigerite]

I agree. I made a big mistake though, the 17.1 should be 18.6% for one, and the win% isn't as low as that, I was calculating for only when SB _and_ BB call, which is wrong. Ugh.

Edit; Or maybe it is 17.1%, this poker calculator thing is a bit on the flakey side with it's %'s, it changes them at will sometimes. Now it's saying 17.1 again.

[tigerite]

[ QUOTE ]
could you provide some clarification on the equation please??:

(0.658) <font color="blue">{win uncontested %}</font> * (450)<font color="blue">{amount won}</font> +
(0.342) <font color="blue">{% called}</font>* (0.238) <font color="blue">{% win if called}</font>* ((2300 * 0.5) + (2150 * 0.5)) <font color="blue">{can you explain this part of the equation} </font>+
(0.342) <font color="blue">{% called}</font> * (0.762) <font color="blue">{% you lose}</font> * (-2000) <font color="blue">{amount lost}</font> =
296.1 + 181.1061 - 521.208 =
-44.0019

[/ QUOTE ]

I copied it from ilya's post /images/graemlins/wink.gif It's basically the mean of the chips - as you don't know which of SB, or BB, or both will call. But, like I said there's a flaw in my calculations, sorry about that.

[SmileyEH]

Ugh...thats what happens when you make a post right after waking up.

-SmileyEH