Two Plus Two Publishing LLC Two Plus Two Publishing LLC
 

Go Back   Two Plus Two Poker Forums > >

Notices

Poker Theory General poker theory

Reply
 
Thread Tools Display Modes
Old 09-05-2017, 09:07 PM   #1
toebester
newbie
 
Join Date: Feb 2016
Posts: 37
GTORB simulation, why no bluff jams?

Doing a sim on a hand I played a few weeks ago live, with another very competent player. Attached is the link to how GTO Range Builder thinks the river should be played, given the game tree I gave it.

The actual hand + ranges, while interesting, are irrelevant to my question.

Facing a bet of 75, why does GTORB advocate villain jamming only KQ, the nuts? MDF for the jam is 38%, and hero is only calling down 34.4% (only calling down nuts), so surely its a spot for GTORB to throw in some infrequent bluff raises.

Link to GTORB Solution

I am new to the solver's, so any help would be welcome.
toebester is offline   Reply With Quote
Old 09-05-2017, 09:45 PM   #2
Frogman3
adept
 
Join Date: Dec 2008
Posts: 952
Re: GTORB simulation, why no bluff jams?

I guess villain can't raise at all here because his range is too weak compared to hero's . But of course when he has the nuts he can raise but if he tries to bluff raise then he will just lose money because hero's range is too strong . So it goes to an equilibrium where villain only raises the nuts and hero only calls the nuts . Essentially villain could just call KQ and it would be the same EV

Did you click to allow shoves ? Because usually hero will just shove here I think and villain will have to bluffcatch
Frogman3 is offline   Reply With Quote
Old 09-05-2017, 10:08 PM   #3
toebester
newbie
 
Join Date: Feb 2016
Posts: 37
Re: GTORB simulation, why no bluff jams?

Quote:
Originally Posted by Frogman3 View Post
I guess villain can't raise at all here because his range is too weak compared to hero's . But of course when he has the nuts he can raise but if he tries to bluff raise then he will just lose money because hero's range is too strong . So it goes to an equilibrium where villain only raises the nuts and hero only calls the nuts . Essentially villain could just call KQ and it would be the same EV

Did you click to allow shoves ? Because usually hero will just shove here I think and villain will have to bluffcatch
I edited the game tree to not allow hero to shove, and villain to check his entire range to me otr, just so things didnt get too crazy.

What you say would make sense if hero had enough KQ in his bet range to make all bluffs immediately unprofitable, but GTORB says villain only calling down 34% (just KQ), compared to the 38% required to prevent profitable bluffs. If 40% of heros range was KQ, then it would make sense to only jam KQ, as bluffing would never be profitable.
toebester is offline   Reply With Quote
Old 09-08-2017, 07:33 PM   #4
Frogman3
adept
 
Join Date: Dec 2008
Posts: 952
Re: GTORB simulation, why no bluff jams?

I know that mdf is a somewhat outdated concept . If you look at GTO solutions many times a player's range won't meet mdf
This has reached an equilibrium where villain can only shove KQ and ofc then hero only has to call KQ . If villain tries to bluff then hero will alter his strategy and make more money so it won't be gto for villain since he could stop bluffing and improve his overall EV . 38% mdf figure is irrelevant in gto solutions

Hero has more KQ than villain and hero has 87s in his range and then JJ . What's going to happen if villain starts bluffing ? Hero will just call some % of his 87s and make more money

Last edited by Frogman3; 09-08-2017 at 07:46 PM.
Frogman3 is offline   Reply With Quote
Old 09-08-2017, 07:36 PM   #5
Brokenstars
Carpal \'Tunnel
 
Brokenstars's Avatar
 
Join Date: Jul 2012
Location: Check out my PGC!!!
Posts: 9,810
Re: GTORB simulation, why no bluff jams?

I think MDF is a good start. Always. You just need to be able to see the forest through the trees (so to speak).
Brokenstars is offline   Reply With Quote
Old 09-08-2017, 07:48 PM   #6
Brokenstars
Carpal \'Tunnel
 
Brokenstars's Avatar
 
Join Date: Jul 2012
Location: Check out my PGC!!!
Posts: 9,810
Re: GTORB simulation, why no bluff jams?

It's also very likely the constraints you've inputted are not great. I.E. the ranges are not the best and/or the sizing and raising constraints are poor.
Brokenstars is offline   Reply With Quote
Old 09-11-2017, 05:35 PM   #7
veginabob
newbie
 
Join Date: Feb 2011
Posts: 46
Re: GTORB simulation, why no bluff jams?

I think the problem here is you put some parameters to the simulation that limited the betting or ranges in a way that made the bot give a false GTO prediction.

GTO solutions need to have a bluffing range because it turns 0 showdown value hands into profitable bluffs and allows you to extract more value from good holdings, it should be ultra rare for a spot not to have bluffs in GTO play(I believe impossible)

Set up a little dummy toy game and compare value/bluff combos matching up versus each other in this scenario without the software and you can solve this spot that way. Its easy to examine a spot, like 2 ranges versus each other on the river on a certain runout. Its about all the possibilities of play coming up to that spot that is the harder solution.
veginabob is offline   Reply With Quote
Old 09-15-2017, 03:38 AM   #8
swc123
grinder
 
Join Date: Jul 2013
Posts: 425
Re: GTORB simulation, why no bluff jams?

Quote:
Originally Posted by toebester View Post
Doing a sim on a hand I played a few weeks ago live, with another very competent player. Attached is the link to how GTO Range Builder thinks the river should be played, given the game tree I gave it.

The actual hand + ranges, while interesting, are irrelevant to my question.

Facing a bet of 75, why does GTORB advocate villain jamming only KQ, the nuts? MDF for the jam is 38%, and hero is only calling down 34.4% (only calling down nuts), so surely its a spot for GTORB to throw in some infrequent bluff raises.

Link to GTORB Solution

I am new to the solver's, so any help would be welcome.
The issue is blockers. The 34.4% is the frequency with which the hero calls when the villain shoves the range that the GTO strategy plays. This means that villain is only shoving KQ and with that hand he still gets call 34.4%. If you look at the % of the hero's range that is KQ at that point in time GTORB says it is 2.78 * 16 = 44.48% whic his the frequency with which hero would call if vilain shoved a hand with no blockers.

GTORB shows the action frequencies that you would observe if you were to imagine watching two players play the GTO strategies against each other. In a situation with blockers the observed frequencies that you would see might not match the frequencies that you would face if you deviated from GTO and bluffed with ATC.

As an extreme example, imagine two players playing a shove/fold HU game 1000bbs deep (this is solved in the MOP. In this case the GTO strategy is to only jam AA and only call with AA.

The observed calling frequency that you would see in this game would be extremely low, because when a shove occured it would block 5 of the 6 AA combos. However, if you were to try shoving a different hand link KK you would get called 6x as often as the observed calling frequency when only AA is shoved.

Blockers + GTO can be super tricky, but hopefully, this makes sense.

-swc
swc123 is offline   Reply With Quote

Reply
      

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


Forum Jump


All times are GMT -4. The time now is 04:32 PM.


Powered by vBulletin®
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © 2008-2017, Two Plus Two Interactive
 
 
Poker Players - Streaming Live Online