Friday, October 16, 2015

Using Bulk GTO Calculations and EV aggregation to improve preflop ranges

Today I'm going to build upon some of the work I did both in my previous post on using bulk GTO calcs to make statistical inferences and on some of the results from my OOP C-betting strategy pack to show how to use GTO postflop calcs to improve a preflop range, even in a preflop spot that is not 2-handed.  I'm going to do this using SimplePostflop's new EV aggregation feature which will automatically solve a given postflop scenario on every possible unqiue flop, then weight those results by how often each flop occurs, and present average EVs for making a preflop call with each combo in the hand range.  Our goal is going to be to use these calculations to work backwards from the average postflop EVs to create a stronger preflop range.

The key concept that we'll need to apply is an idea that I call marginal range building, and I'm going to take a look at how we can use this concept to iteratively strengthen an existing preflop range.  As in my OOP C-betting strategy pack I'm going to be looking at improving a flatting range for a Button caller against an UTG preflop 3x open.  I'm going to start with a range that is already very solid that is used by a student of mine who plays mid/high stakes 6-max on Stars and work on finding a range with an identical number of combos that will have a higher average postflop EV when we call.  The same method can be used to see if we can profitably widen our flatting range by calling a range that includes more combos overall.

The final range that I end up with only differs on about 10 hand combos, but averages about 1.5bb/100 more profit than the initial range, It is played very similarly on almost all flops, it just has slightly more diversified board coverage which yields a higher average EV.

I go through all of this in the video below, and at the bottom of this post I've included links to all of the SPF files I used to do this analysis (10,000+ flop calculations!) for you guys to pour over on your own.  Enjoy!

You can download the solution files here:

If you want to run your own calculations you will need the desktop version of SimplePostflop, its a bit pricey but GTORB readers can get $70 off: here.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.