Second Pooh Bah: How to play poker

[Sushiglutton]

Since my first Pooh-Bah was such a disaster I felt I should try to make a better one. Here are some of the ideas I've had lately. Enjoy!



Making a decission:

<font color="red"> Step 1: Estimate your oppoents strategy. </font>
<font color="green"> Step 2: Calculate your EV for all possible lines. </font>
<font color="blue"> Step 3: Make meta-game considerations. </font>


<font color="red"> Step 1: Estimate your oppoents strategy. </font>

You can devide this in two parts.
i) At the point of your decisson estimate your opponents range
(a range is a list of possible holdings multiplied by their likelyhood.
For example: 0.25*J [img]/images/graemlins/heart.gif[/img]J [img]/images/graemlins/diamond.gif[/img], 0.25*A [img]/images/graemlins/spade.gif[/img]K [img]/images/graemlins/club.gif[/img], 0.1*6 [img]/images/graemlins/heart.gif[/img]7 [img]/images/graemlins/heart.gif[/img],...}.
In theory all possible holdings should be listed)

ii) Estimate how your opponent will play every holding in his range for the rest of the hand, depending on future cards. The actions should be weighted by possibility.

For example: A [img]/images/graemlins/spade.gif[/img]K [img]/images/graemlins/club.gif[/img] Turn 2 [img]/images/graemlins/spade.gif[/img] River 3 [img]/images/graemlins/club.gif[/img] : 0.5 turn b/c, river c/C etc...



<font color="green"> Step 2: Calculate your EV for all possible lines. </font>

Knowing your opps entire strategy it's a straightforward math problem (allthough difficult in practice) to calculate yor EV for all possible lines.



<font color="blue"> Step 3: Make meta-game considerations. </font>

It's not always best to make the action with the momentarily highest EV. You also need to think about how your action in this hand will affect the action in the future. Typical example is making a loose call to make opp play more straightforward in the future.




Ok, now we know how to play poker. There is only one problem. All the above steps are absolutely impossible to complete in real life. Specifically we have the following problem (and many more) for the different steps:





<font color="red"> Problems with Step 1: </font>

i) We do not have enough time.
At the table you have like 5-30s to make a decission. You don't have time to even name the different hands, much less estimate how likely your opp is to hold them. There are ~2500 possibilities for turn and river. No way you can go through all of them in a couple of seconds.

ii) We have not enough information about our opp to make an accurate strategy estimate. There are millions of different things that can happen in holdem. Most of the time we have only seen 0-100 hands that our opp has played.


<font color="green"> Problems with Step 2:
</font>


i) Lack of time.

(in theory this step is very easy, once we have completed step 2)


<font color="blue"> Problems with Step 3:
</font>
i) Lack of time.

ii) Hard to quantify.

iii) Very hard in theory (not yet solved).



What should we do? My idea is that we should try to improve our skills in all three steps, despite the fact that we will never ever get close to be able to complete them. Here are some ideas of how to do that.





<font color="red"> Ideas how to improve ur skills on step 1: </font>


1) Learn to identify the interesting groups of hands.
Depending on the board and action you should break your opps range down in groups (such as TP, FD etc) rather than specific hands (pretty obvious). By analyzing hands and read what other think you should be able to get a grasp of what typical groups are interesting. Also it's important that you try to verbalize this when you are PLAYING. Talk to urself (loud if necessary). 'I think my opp have a weak nine, a gut-shot etc.' Be as specific as you have time to.
Also give the reasons: 'I don't think he holds AA because he didn't 3-bet pf' etc.

2) Play a single table 6max.
This is the best setting for learning reads IMO. 6max is better than FR because you have fewer opps to cocentrate on. Also it's too boring to play a single game FR IMO.

3) Go trough the action again after a hand is played
This is tough, but try to go trough what happened in the hand just played. Verbalize this aswell. Try to draw at least one conclussion about your opps tendencies from every hand.

4) Use the time on obvious decissions to think about ranges. Say you have JJ and a guy to your right raises. You (normally) have an easy 3-bet. So use the extra time you have to spell out a first estimate for opps range.

5) Baby-sit a villain
After a long session (so you have many hands on a villain) replay the session from villain perspective. Think about what he may have been thinking (or not thinking) when he made his moves. What can he have been holding in the hands hedidn't show? I think it may be good to try to understand the 'logic' used by some of our opps.



<font color="green"> Ideas how to improve ur skills on step 2: </font>


1) You can <font color="red"> NOT </font> do this by playing a lot.
This may be suprising, but is very important. Playng a lot is the most important thing to do to improve your skills at step 1 and 2 but it won't help you on this step. In fact it may hurt you. Suppose you have done your best estimate of ur opps strategy and chose an action based on that. Suppose that action is 'wrong' given what villain had. But that has no impact what so ever on how you should do step 2. What you may need to do better is step 1.

2) Deep analyzing hands
Take the time every now and then to make a more serious analysis of a hand. That is give your opponent a rough strategy and calculate your EV for a couple of reasonable lines. Tips is to only do hands where you are HU post-flop. In the start you can do only the river. Then maybe turn+river. Don't try to be to precise, or it will take too long time. Why is this good? Because when you see the numbers and equations you will see the numerical value of differen ideas (for example: What s the cost to give a FC in this spot? Can I make it up later? You will develop a feel for how aggro/tight villain will need to be for you to take a certain line. For example on the river you are often chosing between b/f, b/c, c/c. Agaisnt what type of range/strategy should u take what action? By making many deep analysis you will get a feel you can use in the heat of action.


3) Run computer simulations
This is an idea I had for some time. Actually I have allready discussed this pretty much. The idea is basically the same as 2). But analyzing takes too long time with long boring calculations. For a programmer, they should be pretty easy to automize. Then you can make many more deep analysis. Also you should be able to find break-even points. Meaning, what is the break-even stratey for opp
before this action becomes better than an other one? For now there exists no good poker simulations software. The best software on the market is still this one Sushis river analyzer [img]/images/graemlins/grin.gif[/img] [img]/images/graemlins/grin.gif[/img]!!!


4) Post and discuss in the forum, read books
These are the classical ways to do it. 'Approximate' the calculations by using the classical poker concepts. The problem I have with this is that the concepts are only qualitative not quantitative. Meaning if two concepts point in different directions we are quite lost.

5) POT-SIZE
I'm making a special note on this because it is so extremely important. When you run calculations you will
notice how important this factor is. Always consider pot-size!!!


<font color="blue"> Ideas how to improve ur skills on step 3: </font>


1) Coaching?
To improve on this step I think watching videos and getting a coach may be good. (I have not signed up for anything yet, feeling a bit unsure). Basically this step has a lot to do with frequency: How often do you need to CR on air/ slow-play etc. How should you balance a strategy? The good news is that when you play as low as I do, this is not all that important. I'm still downin value-town, meaning value, not tricky play is the most important concern. I think watching the great players play a session is the best way to get a feel for what a winning style may look like.

[jmitchell42]

nh

[sharpie]

Nice one. Now if only I was good at math and not lazy...

[maverickai]

Sushi, you never mentioned taking part in session reviews for step 3?

[I R BABOON]

[ QUOTE ]

Ideas how to improve ur skills on step 2:


1) You can NOT do this by playing a lot.
This may be suprising, but is very important. Playng a lot is the most important thing to do to improve your skills at step 1 and 2 but it won't help you on this step.

[/ QUOTE ]
You say it's the most important thing for step 1+2 but not for this step which is 2? [img]/images/graemlins/confused.gif[/img] If this isn't a typo, please explain where my reading failed.

[Sushiglutton]

[ QUOTE ]
[ QUOTE ]

Ideas how to improve ur skills on step 2:


1) You can NOT do this by playing a lot.
This may be suprising, but is very important. Playng a lot is the most important thing to do to improve your skills at step 1 and 2 but it won't help you on this step.

[/ QUOTE ]
You say it's the most important thing for step 1+2 but not for this step which is 2? [img]/images/graemlins/confused.gif[/img] If this isn't a typo, please explain where my reading failed.

[/ QUOTE ]

Sorry typo. Suppose to be step 1 and 3

[Sushiglutton]

[ QUOTE ]
Sushi, you never mentioned taking part in session reviews for step 3?

[/ QUOTE ]

Forgot about that. I think you are correct, session reviews are the cheap way to get a coach.

[Sushiglutton]

[ QUOTE ]
Nice one. Now if only I was good at math and not lazy...

[/ QUOTE ]

The lack of time is certainly an issue even off the table...

[Oink]

[ QUOTE ]
Nice one. Now if only I was good at math and not lazy...

[/ QUOTE ]

Sharpie

If you are lazy and bad at math I really wouldnt wanna play you if you were good and had work ethics.