99 In blind steal situation

[elindauer]

Hi Kit,

Again, we need to think about our range on the river. In my last post about handling the 3-bet, I gave us a turn raising range of:

{55,88+,AT,KTs, QTs, JTs, T8s+,A8, J9s, 97d,76d,QJd,Q9d,AKd,AQd,AJd}

which has 67 made hands, and 11 draws.

If a diamond comes in... well, let's see what the range calculator says about our range on this 4d river...

There are 78 combinations in this range on this board.

quads: 1 combo (1.3%)
TT(1)

full house: 8 combos (10.3%)
T8(2) 88(3) 55(3)

flush: 8 combos (10.3%)
AK(1) AQ(1) AJ(1) QJ(1) Q9(1) J9(1) 97(1) 76(1)

trips: 16 combos (20.5%)
AT(8) KT(2) QT(2) JT(2) T9(2)

two pair: 42 combos (53.8%)
AA(6) A8(12) KK(6) QQ(6) JJ(6) 99(6)

one pair: 3 combos (3.8%)
o no hand: 3 combos (3.8%)
J9(3)

Villain is getting 7.5:1 to bluff, and if he has managed his range correctly (and by default we assume he has), he'll still hold combinations that have missed on a diamond river which he can bluff.

So we can fold the worst 13% of our range. Which hands are those?

It's not as easy as just going to the bottom 13%, because we have to remember that we'll be bluff-raising some hands here. Let's figure out what those are:

Let's say we are value raising quads(1), fullhouses(8), and flushes(8). Villain will be getting 8.5:1 to call this raise, so he is indifferent if we are bluff-raising 2 combos as well. So we are bluff-raising our missed J9s hands except for, say, Jh9h.

ok, NOW we can look for the bottom 13% to fold. 13% of 78 = 10 combinations. A8, with 12 combos, is our weakest made hand, so we fold it 5 times out of 6, call everything else.

So... crying call the 99 facing a river donk on this board by default. Folding 99 (and the other 2 A8 combos) would take us to folding 23% of our river range. Exploitable, yes, but perhaps reasonable if we had a read that opponent was passive. Let's check the read...

"Villain is laggy... perhaps FPS... could be on tilt..."

No, this is not a good player to make exploitable folds. If anything, you would adjust to this read by calling the A8 as well and hoping for a busted straight draw. You could even bluff-raise the final missed J9 combo in an effort to take advantage of his range being draw / bluff heavy.

In fact, I think that would be a very reasonable move against this player.


Conclussion: call a river donk with 99.


good luck.
Eric

[James.]

[ QUOTE ]
Hi Kit,

Again, we need to think about our range on the river. In my last post about handling the 3-bet, I gave us a turn raising range of:

{55,88+,AT,KTs, QTs, JTs, T8s+,A8, J9s, 97d,76d,QJd,Q9d,AKd,AQd,AJd}

which has 67 made hands, and 11 draws.

If a diamond comes in... well, let's see what the range calculator says about our range on this 4d river...

There are 78 combinations in this range on this board.

quads: 1 combo (1.3%)
TT(1)

full house: 8 combos (10.3%)
T8(2) 88(3) 55(3)

flush: 8 combos (10.3%)
AK(1) AQ(1) AJ(1) QJ(1) Q9(1) J9(1) 97(1) 76(1)

trips: 16 combos (20.5%)
AT(8) KT(2) QT(2) JT(2) T9(2)

two pair: 42 combos (53.8%)
AA(6) A8(12) KK(6) QQ(6) JJ(6) 99(6)

one pair: 3 combos (3.8%)
o no hand: 3 combos (3.8%)
J9(3)

Villain is getting 7.5:1 to bluff, and if he has managed his range correctly (and by default we assume he has), he'll still hold combinations that have missed on a diamond river which he can bluff.

So we can fold the worst 13% of our range. Which hands are those?

It's not as easy as just going to the bottom 13%, because we have to remember that we'll be bluff-raising some hands here. Let's figure out what those are:

Let's say we are value raising quads(1), fullhouses(8), and flushes(8). Villain will be getting 8.5:1 to call this raise, so he is indifferent if we are bluff-raising 2 combos as well. So we are bluff-raising our missed J9s hands except for, say, Jh9h.

ok, NOW we can look for the bottom 13% to fold. 13% of 78 = 10 combinations. A8, with 12 combos, is our weakest made hand, so we fold it 5 times out of 6, call everything else.

So... crying call the 99 facing a river donk on this board by default. Folding 99 (and the other 2 A8 combos) would take us to folding 23% of our river range. Exploitable, yes, but perhaps reasonable if we had a read that opponent was passive. Let's check the read...

"Villain is laggy... perhaps FPS... could be on tilt..."

No, this is not a good player to make exploitable folds. If anything, you would adjust to this read by calling the A8 as well and hoping for a busted straight draw. You could even bluff-raise the final missed J9 combo in an effort to take advantage of his range being draw / bluff heavy.

In fact, I think that would be a very reasonable move against this player.


Conclussion: call a river donk with 99.


good luck.
Eric

[/ QUOTE ]

phenomenal post.

[One Outer]

I'm not sure I'm capable of doing things like this. Maybe trying to get good at this is a fools errand.

I would appreciate it if someone would put into plain English what the operative concepts are in lin's posts. I think I get it but I'm not sure.

[elindauer]

Hmm... thought I replied to this. Guess it's lost. Sad.

Quick answer: yes there's merit, but you're playing exploitively and you have to be aware of what tendency you are exploiting with this play. That tendency is basically a) won't attack you with lots of junk, just good draws and made hands, and b) frequently bluffs his draws all the way.

Proof left to the reader. I already did it once. :P

good luck.
Eric

[elindauer]

[ QUOTE ]
I'm not sure I'm capable of doing things like this. Maybe trying to get good at this is a fools errand.

[/ QUOTE ]

It's worth the effort. You use this same analysis to play all poker games.

[ QUOTE ]
I would appreciate it if someone would put into plain English what the operative concepts are in lin's posts. I think I get it but I'm not sure.

[/ QUOTE ]

I'm basically talking about techniques for finding game-theory-perfect, unexploitable plays for your entire range. You use these plays as the "default" plays, and vary from there based on player reads.

In my post(s), I use a just a couple key skills to come up with the answer(s). Here are the two most important:

Key skill #1: deciding on a continuation range
decide on what range of hands you are going to play, or your "continuation range". You do this by assesing the odds your opponent was getting to bluff, and then continue with enough hands that he can't profit by bluffing everything. Write one. Buy one. Find one. Just get it done.

Example: to decide that you must call the 99 on the flop, I started by assessing villain's odds to bluff. He's risking 2 small bets to win 5.5, so I absolutely must continue with at least 5.5 of every 7.5 hands in my range. That's top 73%. what hands are those? Write it out by hand, get a tool, whatever. For me, I used my range calculator which produced
these results.

Note that this tells me that I must continue with at least AJ+, so clearly I am not folding 99. There, I have my continuation range, and 99 is in it.

Key skill #2: deciding how to handle that range
Now you want to decide... of the hands in your continuation range, how many should you raise? Key points to keep in mind are that your raise will offer the villain certain odds to continue. You can use those odds to add bluffs to your raising range.

For example, let's say I'm trying to decide which hands to 3-bet on the flop. Well, my raise will offer villain 9.5:1 to call, so for every 9.5 hands I raise for value, I can add 1 bluff.

When choosing hands to raise from your continuation range, you want to keep in mind the following mistakes: Raise too few hands, and villain exploits you by semibluffing / bluffing more. Raise too many, and you are putting in money as an equity underdog against villain's range. Raise the wrong mix, and you skew your range so that you are exploitable in certain scenarios.

You can use math to find out the answer in heads up situations, but I go mostly by feel (for lack of a better alternative right now). I use something like the following rules:

flop: tend to just call. If you do raise, make your raising range {best, middle, some semi-bluff, a bit of bluff} and your calling range {strong-but-not-best, weak, draw}. This seems to raise a good number of hands on the flop while still leaving you a threat to pop it on the turn.

turn: raise most hands with an equity edge, some semibluffs, and a few bluffs. Call with weak made hands and some draws.

river: Usually call weak made hands, fold draws.

This is a really rough blueprint of course, but it gives you an idea.



******
So... that's it. Use a range calculator (perhaps a pencil and paper [img]/images/graemlins/smile.gif[/img]) to find your continuation range. Use a bit of art to find a reasonable way to play that entire range. Use any player reads to alter your default plays a bit, depending on the strength of the read.

Now look at your cards. Which group do they fall into? Make that play. Repeat.



good luck.
Eric

[One Outer]

Thank you

[Mitke]

This is deep stuff. Thanks for all involved and esp elindauer.

This doesn't open up easily or quickly. This isn't necessarily easy to apply on a one hand in isolation (it sure looks laborious) nor at the table. At first sight.

I thought of procedural programming (hand vs hand) vs object-oriented programming (range vs range) when I read this. This involves a similar conceptual leap in your thinking that surely must be worth it.

I vote this thread for the next digest.

[One Outer]

[ QUOTE ]
I vote this thread for the next digest.

[/ QUOTE ]

I agree. I wish I had something to do with making it digest-worthy, but hey, life ain't perfect.

[One Outer]

[ QUOTE ]
Poker is a battle between ranges, not hands. Let's ignore your hand for a minute and look at your range. You didn't tell us what your range is at any point in the hand, so I'm going to have to guess. Please correct me if I guess wrong, and we can fine-tune the numbers.


Preflop: {33+,A4s+,A7+,K7s+,KJ+,Q9s+,QJ,J9s+,T8s+,98s,87s}
stox' opening range from the hijack

[/ QUOTE ]

I think the thread is better if we adjust the range. I'm playing temporarily on a short roll (200 BB or so) so I'm making a concerted effort to decrease my variance. As such, I've narrowed my lp open range.

33+, JTo+, A9o+, K9s, Q9s, J9s

Perhaps this will give us more useful results.

[elindauer]

[ QUOTE ]
This doesn't open up easily or quickly. This isn't necessarily easy to apply on a one hand in isolation (it sure looks laborious) nor at the table. At first sight.

[/ QUOTE ]

No, it's not. There's a very steep learning curve here. I've found that doing it a few times makes it much easier.

It's like anything else in poker... the really good stuff isn't easy. You have to work at it. Remember how hard it was just to keep track of the action when you started playing? Or when you first multi-tabled? Or when you first started thinking about what hands your opponent could hold?

This is just like that.

-Eric