Open Side Menu Go to the Top
Register
Is this a new idea? Worth discussion? Is this a new idea? Worth discussion?

12-27-2018 , 04:46 PM
Quote:
Originally Posted by RustyBrooks
Well ok but your problem is that you have a misconception about how ranges are constructed. And when you discovered your mistake you seem to have decided it's a flaw in how people think about ranges, except for the most part, people don't make the same mistake you did. I think it's a mistake that would mostly be made by people writing simulators without thinking too much about range construction.

If you want to raise X% and call Y% then you don't use "top y%" for your calling range, you use "top x+y%, minus top x%" or put another way, you start your calling range where your raising range ends.
That is another way to say what I said. Some ranges must consider other ranges. By excluding hands in the raising range from the calling range then calculations using combos will be correct. A good idea if you use a tool like Holdem Manages.

I have spent a lot of time calculating the EV, not equity or opinion, for a full game. It is a major focus.
Is this a new idea? Worth discussion? Quote
12-27-2018 , 04:47 PM
Quote:
Originally Posted by outfit
Are you using an object oriented language?
Java. So yes.
Is this a new idea? Worth discussion? Quote
12-29-2018 , 01:43 PM
I still can't understand what the hell this is all about.
Is this a new idea? Worth discussion? Quote
12-29-2018 , 02:14 PM
Quote:
Originally Posted by Iblis
I still can't understand what the hell this is all about.
Puking on elegance
Is this a new idea? Worth discussion? Quote
12-30-2018 , 12:12 AM
Quote:
Originally Posted by Fishing
I have not suggested that there is a separate shadow range, only that some ranges are in the shadow of other ranges.

Using combos to calculate range percentages will result in incorrect values if the same hand is in a range that shadows another range.

There are many other ranges that are in the shadow of another range. If you have an Isolate or Steal range preflop, then they also have are shadowed by other ranges.

For example, it is possible to have a Flop MDF range. Such a range can only include hands that you continued with preflop. For a half pot bet the MDF is 66%. If the preflop hands that form the MDF range are reduced to 66% then you have an accurate way to implement MDF. Whatever rules that you use to call or bet draws and made hands on the Flop have to be considered first, before the MDF range. There are also other considerations. But the basic idea is to have a simple way to implement MDF ranges that are accurate, including the hands with the highest EV, and include randomness.

I am developing a full game simulator, not a Bot. The simulator is for a full table, not a limited subset. After a year developing the code, I have learned a few things that might be useful to others and want to share what I have learned.

I started looking at shadow effects because the Simulator produced some unexpected results. A 10% Call range might result in only 8% of actual calls.

Arty is correct in that the simulator, is a linear process. When I play I also use a linear process with great results. EV first.
It just sounds like you started out not knowing how to correctly construct a range and have now realised your mistake. Welcome to 2018.
Is this a new idea? Worth discussion? Quote
12-30-2018 , 04:44 PM
Quote:
Originally Posted by BornToRun
It just sounds like you started out not knowing how to correctly construct a range and have now realised your mistake. Welcome to 2018.
True. Now that I understand better, I am sharing that information.
Is this a new idea? Worth discussion? Quote
12-30-2018 , 06:28 PM
Quote:
Originally Posted by Fishing
True. Now that I understand better, I am sharing that information.
Keep doing what you are doing. It is cool. In was just being an ass hole because I was jealous how many posts you're thread got. Keep the ideas simple. So they make sense even to someone that does not play poker. And maybe share you're code.
Is this a new idea? Worth discussion? Quote
01-01-2019 , 03:35 PM
Quote:
Originally Posted by outfit
Keep doing what you are doing. It is cool. In was just being an ass hole because I was jealous how many posts you're thread got. Keep the ideas simple. So they make sense even to someone that does not play poker. And maybe share you're code.
I will not share the source code, but the functionality will be available in an application Forum rules do not allow me to talk about the application. Sorry.
Is this a new idea? Worth discussion? Quote

      
m