Open Side Menu Go to the Top
Register
Expected Value Expected Value

05-28-2017 , 10:43 PM
How do you quickly calculate EV while in a hand. Also, when and how do you use EV. Examples?

Thanks
Expected Value Quote
05-29-2017 , 09:27 AM
Quote:
Originally Posted by TheShakeDaddy
How do you quickly calculate EV while in a hand. Also, when and how do you use EV. Examples?

Thanks
EV is a tool used by economist and gamblers alike to determine the value of an action if you were to perform the exact same action a larger number of times. This is done creating a weighted sum of the outcomes, and then subtracting the cost.

For example, if I have a 10 sided dice, it costs me $1 to roll the dice, and I get paid $11 if I roll a 10, then I calculate EV as follows
EV=(the sum of Probability of an Outcome*The value of the outcome)-cost of the action

So
EV=(10%*$11)+(90%*$0)-($1). EV in this game is $.1. If you do this game a million times, you will average a net value of $.1 per bet.

Anytime EV is positive, it is a good action to take. If it is negative, it is a losing action

to calculate EV, you have to create an outcome tree, and determine the liklihood and value of each outcome. This can be exceedingly diffcicult in poker when there are subsequent rounds of betting and stacks are deep.

For example, let's say I call a preflop raise of 3BB from the big blind with QTo. The flop comes KJ3r. I check, and the villain bets 5 BB. If I call, you have the following outcomes

1. I make my straight, and can get more action from villain
2. I make my straight, but villain shuts down
3. I don't make my straight on the turn, but villain doe not bet
4. I don't make my straight, and villain bet
5. I don't make my straight on turn or river, and I win at showdown
6. I don't make my straight on turn or river, and i lose at showdown.

And there are obviously a lot of other branches from the outcome tree that can occur on the turn and river.

This becomes almost impossible to calculate on the fly. What I will do is simplify and act is if I am playing with one card to go. I look at the value of the call, the likelihood of making my hand and how much I win and the likelihood of missing my hand. If I am looking to make a bet, I will also look at the likelihood of the villain folding (keeping in mind the amount you win if the villain folds is less than if he calls and you make your hand)

To estimate EV accurately, you need to be able to range your opponent, estimate how likely it is that he folds in certain situations, and estimate your likelihood of hitting certain hands. I usually practice this when I am driving out to the casino, and also practice this a ton using an equity calculator and Excel.

Note: my EV formula differs slightly from a lot of people. The use the formula
EV=(probability you will win*the pot you will win not including your call\bet)-(probability you will lose*the amount of the call\bet). It amounts to the same, I just find it easier to visualize if you place all the action from a round into the pot.
Expected Value Quote
05-29-2017 , 10:08 AM
Quote:
Originally Posted by SpewingIsMyMove
EV is a tool used by economist and gamblers alike to determine the value of an action if you were to perform the exact same action a larger number of times. This is done creating a weighted sum of the outcomes, and then subtracting the cost.



For example, if I have a 10 sided dice, it costs me $1 to roll the dice, and I get paid $11 if I roll a 10, then I calculate EV as follows

EV=(the sum of Probability of an Outcome*The value of the outcome)-cost of the action



So

EV=(10%*$11)+(90%*$0)-($1). EV in this game is $.1. If you do this game a million times, you will average a net value of $.1 per bet.



Anytime EV is positive, it is a good action to take. If it is negative, it is a losing action



to calculate EV, you have to create an outcome tree, and determine the liklihood and value of each outcome. This can be exceedingly diffcicult in poker when there are subsequent rounds of betting and stacks are deep.



For example, let's say I call a preflop raise of 3BB from the big blind with QTo. The flop comes KJ3r. I check, and the villain bets 5 BB. If I call, you have the following outcomes



1. I make my straight, and can get more action from villain

2. I make my straight, but villain shuts down

3. I don't make my straight on the turn, but villain doe not bet

4. I don't make my straight, and villain bet

5. I don't make my straight on turn or river, and I win at showdown

6. I don't make my straight on turn or river, and i lose at showdown.



And there are obviously a lot of other branches from the outcome tree that can occur on the turn and river.



This becomes almost impossible to calculate on the fly. What I will do is simplify and act is if I am playing with one card to go. I look at the value of the call, the likelihood of making my hand and how much I win and the likelihood of missing my hand. If I am looking to make a bet, I will also look at the likelihood of the villain folding (keeping in mind the amount you win if the villain folds is less than if he calls and you make your hand)



To estimate EV accurately, you need to be able to range your opponent, estimate how likely it is that he folds in certain situations, and estimate your likelihood of hitting certain hands. I usually practice this when I am driving out to the casino, and also practice this a ton using an equity calculator and Excel.



Note: my EV formula differs slightly from a lot of people. The use the formula

EV=(probability you will win*the pot you will win not including your call\bet)-(probability you will lose*the amount of the call\bet). It amounts to the same, I just find it easier to visualize if you place all the action from a round into the pot.


This is great!!!!
Expected Value Quote
05-29-2017 , 01:47 PM
Quote:
Originally Posted by SpewingIsMyMove
EV is a tool used by economist and gamblers alike to determine the value of an action if you were to perform the exact same action a larger number of times. This is done creating a weighted sum of the outcomes, and then subtracting the cost.

For example, if I have a 10 sided dice, it costs me $1 to roll the dice, and I get paid $11 if I roll a 10, then I calculate EV as follows
EV=(the sum of Probability of an Outcome*The value of the outcome)-cost of the action

So
EV=(10%*$11)+(90%*$0)-($1). EV in this game is $.1. If you do this game a million times, you will average a net value of $.1 per bet.

Anytime EV is positive, it is a good action to take. If it is negative, it is a losing action

to calculate EV, you have to create an outcome tree, and determine the liklihood and value of each outcome. This can be exceedingly diffcicult in poker when there are subsequent rounds of betting and stacks are deep.

For example, let's say I call a preflop raise of 3BB from the big blind with QTo. The flop comes KJ3r. I check, and the villain bets 5 BB. If I call, you have the following outcomes

1. I make my straight, and can get more action from villain
2. I make my straight, but villain shuts down
3. I don't make my straight on the turn, but villain doe not bet
4. I don't make my straight, and villain bet
5. I don't make my straight on turn or river, and I win at showdown
6. I don't make my straight on turn or river, and i lose at showdown.

And there are obviously a lot of other branches from the outcome tree that can occur on the turn and river.

This becomes almost impossible to calculate on the fly. What I will do is simplify and act is if I am playing with one card to go. I look at the value of the call, the likelihood of making my hand and how much I win and the likelihood of missing my hand. If I am looking to make a bet, I will also look at the likelihood of the villain folding (keeping in mind the amount you win if the villain folds is less than if he calls and you make your hand)

To estimate EV accurately, you need to be able to range your opponent, estimate how likely it is that he folds in certain situations, and estimate your likelihood of hitting certain hands. I usually practice this when I am driving out to the casino, and also practice this a ton using an equity calculator and Excel.

Note: my EV formula differs slightly from a lot of people. The use the formula
EV=(probability you will win*the pot you will win not including your call\bet)-(probability you will lose*the amount of the call\bet). It amounts to the same, I just find it easier to visualize if you place all the action from a round into the pot.
Hi Guys, first of all as this is my first post I would like to say that I'm happy to join this community and I'm looking forward to share an ideas and be a part of tough poker brainstorms

SpewingIsMyMove, very nice post, can you tell sth more about analizing villian's range please? I mean what tools do you use? How often are you right ? Im interesting in GTO and started reading a lot about this and honestly I would like to improve my opponents range assessment.

Have a good day all
Expected Value Quote
05-29-2017 , 05:42 PM
Quote:
Originally Posted by RunGoodPlzzz
Hi Guys, first of all as this is my first post I would like to say that I'm happy to join this community and I'm looking forward to share an ideas and be a part of tough poker brainstorms

SpewingIsMyMove, very nice post, can you tell sth more about analizing villian's range please? I mean what tools do you use? How often are you right ? Im interesting in GTO and started reading a lot about this and honestly I would like to improve my opponents range assessment.

Have a good day all
Analyzing ranges is the tricky part, and I would be lying if I told you that I don't get bit in the ass when a player turns over the nuts with cards I had eliminated from his range.

I am not the best at narrowing ranges (others here are much better) but I will tell you what factors I consider

1. Experience level or skill level of a player
2. Playing style of a player
3. Mental state of player
4. Immediate context of player (for example, is this the third hand he has opened in a row)
5. Position of a player
6. Make up of a player
7. Any personal dynamic that player has with the table
8. Overall game context (I play mostly tournament, so amount of time left in a level, proximity to the bubble, etc. are all factors)
9. Action before a player
10. Image of players remaining to act

For example, one of the easiest ranging problems you will encounter is in a tournament, a player is down to about 4 to 10 BB, and he is approaching middle position. At this point, pretty reliably, most players will open their range up for unopened pots and shove with Ax, K9+, QT+, and any pocket pair. If I have as big as AT or KQ, I am pretty much going to call.


That being said, my understanding of GTO is that is specifically does not require reads and range narrowing to function (I have only dabbled in GTO, but have come to understand that it is not terribly useful at the lower to middle skill levels where players are more unbalanced and exploitable).
Expected Value Quote

      
m