Tuesday, April 22, 2008

Blind defense

I feel like I fold my blinds too much to steal-happy players on the button, and I'm not going to take it anymore.

In an effort to stop these over-aggressive players always try to rob my blinds, I'm going to attempt to construct a blind defense range. I'll want 55 percent equity against my opponent's range in order to neutralize his positional advantage.

One difficulty is that it's difficult to determine what an opponent who steals X percent of the time is like based on PokerTracker statistics. PT calculates Attempt to Steal percentage using raises from the cutoff or from the button when everyone else in the hand has folded before them. The problem with this calculation is that it results in a lower number than you might expect. I think that's because it takes your steal percentage out of all hands played, not just those where everyone in early positions has folded to you. I still think this exercise might be worthwhile; it's just that these steal figures won't correspond with PT's.

1. Against someone who steals with any two cards:

If I want 55 percent equity against that range, I can defend with my top 65 percent of hands:
22+,A2s+,K2s+,Q2s+,J2s+,T2s+,94s+,84s+,74s+,64s+,54s,A2o+,K2o+,Q4o+,J6o+,T7o+,97o+,86o+,76o

Damn, that's a lot of hands. Any Kx and most Qxo hands, as well as 54s and up.

2. Against someone who steals 50 percent of the time:
22+,A2s+,K2s+,Q2s+,J4s+,T6s+,96s+,86s+,76s,65s,A2o+,K5o+,Q7o+,J7o+,T8o+,98o

I can defend with the top 30 percent of hands:
55+,A2s+,K5s+,Q7s+,J8s+,T8s+,98s,A7o+,A5o,K9o+,Q9o+,J9o+,T9o

3. Against someone who steals 33 percent of the time:
22+,A2s+,K5s+,Q7s+,J8s+,T8s+,98s,A5o+,K8o+,Q9o+,J9o+,T9o

I can defend with the top 20 percent of hands:
66+,A4s+,K8s+,Q9s+,J9s+,T9s,A9o+,KTo+,QTo+,JTo

4. Against someone who steals 20 percent of the time:
66+,A4s+,K8s+,Q9s+,J9s+,T9s,A9o+,KTo+,QTo+,JTo

I can defend with the top 12 percent of hands:
77+,A9s+,KTs+,QTs+,JTs,ATo+,KJo+

All data from PokerStove

10 comments:

Eric a.k.a. Bone Daddy said...

I think there is a lot of value to consistently folding early to a foolish button thief. It gives them a false sense of hope and creates an initial tight image.

I know this is simplistic, but after 4 or 5 orbits of folding, you only lose 5bbs. At this point, I look for the 2 ugliest cards i can find in the sb or bb, and I re-pop him. If he folds, I have by BBs back, and I have to show my crappy cards. If he calls, I lead out on the folp regardless, and again, I mostly get folds as I've been playing tight.

Then it tends to be payday when you finally find a hand in the blinds. It also helps create a "looser" table image, than I am actually playing.

I like the line of thought of a postive range to push back, but creating large pots OOP with marginal holdings sounds like a bad idea.

Gnome said...

Good points. To some extent this analysis is more theoretical than practical, especially since many of the weaker range of blind defense hands are difficult to play postflop.
Your argument about allowing blind thieves some leeway is also backed up by anecdotal evidence:
1. Some of the top players seldom defend their blinds.
2. Some of the biggest losers are those who frequently try squeeze plays and have high blind defense numbers.

Fuel55 said...

Are we talking "flat-call" defending or "3-bet" defending?

Gnome said...

Either way -- flat calling or 3-betting, depending on the situation.

Eric a.k.a. Bone Daddy said...

i actually had to go to blind defense rehab last year. Lots of celebrity poker players there, most are dead now.

Wes said...

The type of player that is going to be liberally raising on the button is probably going to be someone that is going to cause you tons of problem postflop. Defending your blinds more isn't really the answer unless the player is weak postflop.

If you protect more, the player will just 2barrell more if you check/call too much with medium weak pair hands, float/bluff raise you more if you c/raise too much.

Being OOP sucks. A LOT. I would never go by equity preflop when playing no limit. What is that going to accomplish other than showing you are ahead of their range preflop? Answer: nothing.

Greylocks said...

In cash games with no antes, let them have your blinds. I've always felt the proper "blind defense" in cash games is to steal other people's blinds when you're in late position

The issue is of considerably greater importance in tournaments, especially when the antes kick in and/or the stacks are getting short relative to the blinds. Now the blinds are worth a lot more than they are in cash games.

Gnome said...

I think you guys are right. Thanks for the comments.

spritpot said...

I will nth the comments suggesting against loose blind defense. But, IF you decide to defend more often, I'll suggest that the poker stove exercise you're doing isn't very helpful towards figuring out which hands to call with. The reason is that many of the hands that run decently on all-in preflop confrontations, which is essentially what your poker stove numbers measure, play very poorly with 100bb stacks. The ace-rag and king-rag hands come to mind. On the other hand, many of the hands that DO play well post-flop don't run so well all-in preflop, such as 67s. For instance, your #3 here has you defending with A9o but folding 87s, which I think is a mistake. The implied odds of playing 87s and the reverse implied odds of playing A9o far outweigh the all-in pre-flop equity edge A9o has over your villain's range. Poker hands are only valuable in so far as they can extract value from other hands. It's pretty rare to win a big pot with A9o, but I'm sure you've won many with 87s.

bruechips

CC said...

What about the Andy Beal-type defense, where you might defend every x times or when you hold a red face card or something like that? A bit mechanical, but I'd be interested in your thoughts about this. I'm fairly passive with the blinds and defend more based on feel with non-goot hands.