explain to me the mathematics of a blocking bet

ipokeder · #1 08-15-2007, 02:02 PM

the use of "fake blocking bets" is very common in mid-high stakes as bluff-inducing bets, but personally i never use any actual blocking bets in my game and to an extent i feel like i'm not really sure what they are or what purpose they should serve. and since i don't use any real blocking bets, i don't think it makes sense for me to incorporate fake blocking bets against competent opponents, so i'd like for someone to explain to me the mathematics in which a blocking bet (excluding metagame considerations that i can make fake blockers in the future and whatnot) is the most +EV play

assuming the villian is rational and good

say for example, i have 44 w/ 4d on a 3c 6d 7d 8d 2d board and i block-bet the river

or i have JJ in an RR'd pot, the board is Qs Tc 5s 4c Th and i block-bet the river

are these examples of blocking bets? is a blocking bet fundamentally different from a bet/folding hand (because i always consider that a value bet, even if i'm thinly vbing an underpair or something but definitely folding to a raise)

i've always had the intuitive feeling that if i think i'm ahead in a hand i should be check/calling or value betting to get calls from worse hands or if i feel i'm behind i should be check/folding or bluffing to make him fold his hand, but i don't understand where block-betting comes into this (assuming block-betting is not simply bet/folding)

thanks

ipokeder · #2 08-15-2007, 02:14 PM

some ideas that i've had that i'm not sure whether they're right or retarded -

perhaps the purpose of an effective blocking bet is to polarize the range of our opponent's betting range such that the expected value of making any certain decision is also polarized so that worse plays become worse and better plays become better since his range is not as wide as when he has the initiative in the hand, so our average decision becomes more +EV and the variance in the outcome (that is, whether or not we made the right decision) of certain hand is smaller

or perhaps it's very useful in changing the size of the pot without actually making any kind of substantial action because the pot size we currently have would be inconvenient or difficult to play with

yes or no?

xorbie · #3 08-15-2007, 02:25 PM

block bet as a term is retarded and useful only against the most retarded of droolers who for some reason you suspect are going to go all in when you'd rather only put less than that in the pot.

value bets are good though.

ipokeder · #4 08-15-2007, 02:30 PM

well i swear to god in one of his videos sbrugby bet 44 w/ 4d vs a competent player on a board like that on the river OOP and called it a blocking bet

and why would you call it a fake blocker if you never use a real blocket? is it more accurately described as a fake thin vb then?

tcorbin16 · #5 08-15-2007, 02:54 PM

I use the Allin button as a blocking bet

Irishman07 · #6 08-15-2007, 03:23 PM

I think there could be some use for a blocking bet if you utilize fake blocks as well, especially if you're deep. Say for instance you're pretty sure villain has overpair or busted draw, and bluffs optimally so you can't call if he bets. If you bet ~1/2 pot with hands like tpgk, and occasionally a set it really makes it hard for him to raise because he's gotta risk a lot more if you come back over the top. Not sure if this really makes sense, and it gets really complicated because it depends so much on opponent's bluffing frequency. In theory I think it should work but in practice its tough to really gauge because not many people do it. Here's an example of a block bet gone wrong:

Full Tilt Poker No-Limit Hold'em, $4 BB (6 handed) Hand History converter Courtesy of PokerZion.com

saw flop|saw showdown

CO ($108)
Button ($334)
Hero ($400)
BB ($156)
UTG ($770)
MP ($251.40)

Preflop: Hero is SB with Q[img]/images/graemlins/heart.gif[/img], K[img]/images/graemlins/heart.gif[/img].
1 fold, MP raises to $14, 2 folds, Hero calls $12, BB calls $10.

Flop: ($42) T[img]/images/graemlins/diamond.gif[/img], 3[img]/images/graemlins/club.gif[/img], 2[img]/images/graemlins/club.gif[/img] (3 players)
Hero checks, BB checks, MP checks.

Turn: ($42) 8[img]/images/graemlins/club.gif[/img] (3 players)
Hero bets $20, BB folds, MP calls $20.

River: ($82) 3[img]/images/graemlins/diamond.gif[/img] (2 players)
Hero bets $40, MP raises to $124, Hero raises to $366 (All-In), MP calls $93.40 (All-In).

Final Pot: $665.40

Results in white below: 
Hero has Qh Kh (one pair, threes).
MP has As Ks (one pair, threes).
Outcome: MP wins $516.80. Hero wins $148.60. 

On the turn I figured I'd take a cheap stab at the pot and when original raiser called I put him on either overs probably with the flush draw or a mid pp. He was the type that wasn't gonna slowplay a set on the flop or a flush draw + overs. So when I bet the turn I was hoping to take it down but also for information/setting up a river play. I figured my 1/2 pot would be +ev since he probably had marginal pair/overs roughly 50/50 of the time. If he folds his overcards most of the time I'll be good the 33% I need to break even. When he raised though I could put him on a busted draw with pretty good confidence so I shoved. Unfortunately he was apparently a thinking player and we weren't deep so I got owned. But if I ever take that line with a flush against him now I know I can bust him.

Short version: Block bets could probably work, but against good thinking players its gonna be real tough to figure out the optimal strategy.

LearnedfromTV · #7 08-15-2007, 03:50 PM

Labeling a bet a bet/fold vbet or a blockbet is semantics if the bet is the same size, so I'm assuming by a blockbet you mean an abnormally small bet.

$200 pot on the river.

Your options are check, bet $50 (blocking bet), bet $150 (regular value bet).

Your opponent has you beat 35% of the time. If you check, he'll bet $150 45% of the time, bluffing 10%, value betting 35%, never checking behind a better hand. If you bet $150, he'll call or raise with every hand that beats you and he'll call w/ 30% of his worse hands. But if you bet $50 he'll call with 70% of his worse hands.

So check-calling is -EV: $350*(.1/.45) - $150*(.35/.45) = -$38

And betting $150 is -EV: $150*.65*.3 - $150*.35 = -$24

But betting $50 is +EV: $50*.65*.7 - $50*.35 = $5.

So that's the math.

Problem is, this example assumes he never bluff raises. And generally, if he's bluffraising some, he's probably going to bluffraise the $50 more often than the $150, which mitigates the advantage you get from him calling wider.

I generally agree with xorbie (blockbets are usually not the best option except against some fish, or at least against someone who is really predictable in a certain situation), but it seems like a decent time to use such a bet is on a fourflush board with a medium flush, like the 9. Any flush and maybe some other stuff will curiosity call and you get a little value that you wouldn't get from a normal bet. Works as long as you aren't likely to get bluffraised, which, against a decent oppponent, may require occasional fake blockbets with the nuts. I suppose against the right opponent you can fake blockbet a non-nut flush, although that's probably not the kind of guy an actual block would be good against.

Mook · #8 08-15-2007, 04:07 PM

[ QUOTE ]
I generally agree with xorbie (blockbets are usually not the best option except against some fish, or at least against someone who is really predictable in a certain situation), but it seems like a decent time to use such a bet is on a fourflush board with a medium flush, like the 9. Any flush and maybe some other stuff will curiosity call and you get a little value that you wouldn't get from a normal bet. Works as long as you aren't likely to get bluffraised, which, against a decent oppponent, may require occasional fake blockbets with the nuts. I suppose against the right opponent you can fake blockbet a non-nut flush, although that's probably not the kind of guy an actual block would be good against.

[/ QUOTE ]
Anyone who believes the blocking bet is "retarded" doesn't understand the concept.

It is true, however, that in order for a blocking bet to have any value whatsoever, it must come from a range of hands that includes some very strong ones as well. Your goal is to make your opponent indifferent to raising or calling (we're assuming he'll never fold), which means he has to believe when you make the bet that raising is not a +EV play.

If you're holding a mediocre hand (like the 4d above with four diamonds out), the blocking bet may well be -EV in that situation, but it may still be your best play because the alternative - checking and allowing worse hands to check behind and better hands to value bet - is even more -EV. But there has to be a nonzero probability that, whatever bet you're making in that situation with the 4d, you might also make with the Ad. Otherwise, all you're doing is throwing money away in that spot.

As is often the case, The Mathematics of Poker explains the concept far better than I could. Blocking bets are a powerful and IMO underappreciated tactic, especially on the river.

Mook

LearnedfromTV · #9 08-15-2007, 04:18 PM

[ QUOTE ]
the blocking bet may well be -EV in that situation, but it may still be your best play

[/ QUOTE ]

This is impossible. EV(checkfold) = 0

jogsxyz · #10 08-15-2007, 04:36 PM

[ QUOTE ]

As is often the case, The Mathematics of Poker explains the concept far better than I could. Blocking bets are a powerful and IMO underappreciated tactic, especially on the river.
Mook

[/ QUOTE ]

Can't find blocking bet in the index or table of context of
The Mathematics of Poker . There is a chapter on
bet sizing. Can you cite the page from MOP on blocking
bets?
Believe blocking bets was created by a pseudo math guy.