Tuesday, April 03, 2007

ICM for SNGs - 2+2 Article

The 2+2 'Magazine' Articles this month contain something on ICM for Sit and Goes it is called "Playing with the ICM formulas' by Alan Kratz - link to this is below (right click / open in new window please!).


Now it certainly looks impressive, what with all the algebra and everything - but I wanted to ask the question whether this tells us anything new??

Well, no. It does a decent job of explaining the equations (if anything making it seem more complicated than it really is) and suggests you can look at ICM decisions 2-ways around but that is about it...

First thing is how to decide if your call is +$ev, Plan3t Gong readers already know this one right?? Compare your winning chances vs opponents range with the risk / reward in $equity for the subsequent win / loss. So if you feel you have 50% vs opponents range but would risk $25 to win $12.5 more you have a fold and if you have 80% vs opponents range (ie aces!) in the same situation you have a call.

Note: For those who missed them there is a 'jump off page' for ICM articles listed on the right.

Putting in the break even point is useful to some extent in the linked article... you need to decide in advance what % of equity gain you consider big enough to call.

1% is usually considered good here... but this can be influenced by many factors including stack sizes of opponents (particularly very small stacks), perceived skill levels (if you are much better then look for a better spot - if you are out of your depth then take 0.5% edges and be happy!) and whether the blinds will go up next hand.

For the pushing side things are equally straightforward - compare you equity gain from picking up the blinds with the % chance you will be called and your winning chances thereafter. If your winning chances + gain from the get blinds*fold % are positive in comparison to the risk / reward in equity when called then push - if not then fold....

this is why a fold at BB200 can easily become a push at BB400 - the preflop pot is that much bigger that the equity gain for opponents folding outweighs the 'getting called' side of the equation (assuming you have at least some winning chances when you are called.).

What Alan Kratz suggests is that you can approach these situations from both sides;

- What % vs your opponents range do you need given the equity at stake? or
- What equity gain do you need to call given you assessment of your opponents range?

While each has its merits my personal preference is the first method - since stack sizes are there to see it is relatively easy to work out your $ev and make the assessment of what you are risking in any given situation - if you do not yet know this then a suggestion would be to get an ICM calculator and make a chart of 10 or so common stack size combos at the bubble (will only take 20 minutes and is well worth it!!).

Your assessment of your opponents pushing and / or calling ranges is more read dependant (have covered this in previous posts) but tools such as PokerStove give you the basics.

Final suggestion is to get a copy of SNG power tools - plugging in common situations and seeing the mathematically correct push / fold ranges is worth the $80 or so price very quickly.

In conculsion then a decent article on the subject, but too fancy for some principals which are actually quite straightforward!

Cheers, Mark

1 comment:

Short-Stacked Shamus said...

Another good post, Mark. I'm not playing a lot of tourneys at the moment, but I when I do I know I'll be going back through many of these excellent discussions for tips. Keep up the good work.