Open Side Menu Go to the Top
Register
Yeodan's GTO / solver output questions and notes Yeodan's GTO / solver output questions and notes

06-18-2021 , 08:42 AM
I'm gonna have a lot of questions, so I'll just start a thread to group them in!

I'll also try to post the stuff I learn.
Yeodan's GTO / solver output questions and notes Quote
06-18-2021 , 08:48 AM
So I just played this hand.
6-max zoom 2NL on PokerStars EU

I open from MP (HJ)
BB calls, 100bb eff
I cbet flop and he folds

I was wondering about what size to use on the flop, so I ran this with a 58.7% BB range and a simplified GTO HJ range 18.3%

My goal is purely to learn GTO, not how to play against 2NL players.
I realize this player is probably not defending 60%.

Is 60% for the BB ok here? I opened to 2bb.
I know if I open to 2bb on the BTN a GTO defense range would be around 80%


Anyway, I got this output, which is mixing between all my sizings about equally, except the 150% pot bet:

150% pot: 1.71%
100% pot: 14.1%
66% pot: 16%
33% pot: 14.4%
check: 53.8%

This is completely unusable for me, I cannot learn to use 3 bet sizes.
Maybe 2, but I'd really prefer just one.

What do I do with this output? Where do I go from here?

Yeodan's GTO / solver output questions and notes Quote
06-18-2021 , 08:53 AM
Could I just re-run this with just 33% and 66% for the flop?
Those are probably the default sizes I want to learn how to use, except in spots where I can really go smaller or larger.
Yeodan's GTO / solver output questions and notes Quote
06-18-2021 , 11:00 AM
Those ranges are awful my man.
Input preflop solver ranges
Yeodan's GTO / solver output questions and notes Quote
06-18-2021 , 09:35 PM
Ah yeah I forgot to remove top of BB range again lol ...
Don't have solved ranges, feel free to send me $2000 so I can buy some!
Yeodan's GTO / solver output questions and notes Quote
06-19-2021 , 07:43 AM
Quote:
Originally Posted by Yeodan
Ah yeah I forgot to remove top of BB range again lol ...
Don't have solved ranges, feel free to send me $2000 so I can buy some!
you can get zenith for 79 or 179 cant remember what i payed for them for sure. Monker ranges are like 500. Not sure what ranges would cost 2k. Plus i think GTO wizard has a free trial.
Yeodan's GTO / solver output questions and notes Quote
06-19-2021 , 07:45 AM
I have a free range viewer if you want, you can go on my Discord and see the instructions to access it
Yeodan's GTO / solver output questions and notes Quote
06-19-2021 , 01:36 PM
Quote:
I know if I open to 2bb on the BTN a GTO defense range would be around 80%
This is maybe true heads up, but never in 6max games.

Quote:
Could I just re-run this with just 33% and 66% for the flop?
Those are probably the default sizes I want to learn how to use, except in spots where I can really go smaller or larger.
Absolutely you could. When you get a bunch of mixed sizes like this I think it's important to try and find strategically distinct paths. All this sim tells you is that there are many playable strategies. Also, consider learning how to use the different betting tabs and compare how the eV/Equity distribution changes. Maybe pick up Flopzilla for $25 so you can export ranges back and forth for further analysis.

Quote:
Don't have solved ranges, feel free to send me $2000 so I can buy some!
I thought I saw you using the book Modern Poker Theory, that one has plenty of solved ranges. GTO wizard, solver+, and zenith poker have free GTO ranges you can use without buying the full thing.
Yeodan's GTO / solver output questions and notes Quote
06-22-2021 , 02:57 PM
You should also look at the EV difference between different actions.

Many times, when the solver chooses a mixed strategy with a type of hand, its because the EV is basically the same between the actions. In that case, you can choose a single bet size to simplify your strategy without giving up value.
Yeodan's GTO / solver output questions and notes Quote
06-22-2021 , 09:48 PM
Quote:
Originally Posted by ledn
You should also look at the EV difference between different actions.

Many times, when the solver chooses a mixed strategy with a type of hand, its because the EV is basically the same between the actions. In that case, you can choose a single bet size to simplify your strategy without giving up value.
Anytime the solver chooses mixed actions, the EV is the same. That's a condition of equilibrium. Why would you intentionally play a worse strategy?

But be careful. Oversimplifying can make you exploitable. The only reason a solver mixes is to reduce the exploitability of your strategy.
Yeodan's GTO / solver output questions and notes Quote
06-23-2021 , 09:54 AM
Hello.

I recommend you to study the free theory content from Zenith. It's very good.
Yeodan's GTO / solver output questions and notes Quote
07-21-2021 , 11:40 AM
Maybe a silly question, but can GTO (NE) ever be -EV?

Someone posted this in one of my threads in another forum and I replied that it's impossible for GTO to be -EV.
Why would GTO ever take a -EV action?

Obviously GTO could be -EV if it can't beat the rake, but I doubt that applies in any of the games I play.

Is it just plain wrong to say GTO can be -EV?
Yeodan's GTO / solver output questions and notes Quote
07-21-2021 , 12:01 PM
Great question Yeodan.

Its all about perspective. EV is usually measured relative to another action. GTO strategies will never intentionally take an action that's lower EV than other actions. e.g. it won't call if calling is lower EV than folding.

However, net EV can definitely be negative. A table of GTO bots at a raked table will have net -EV, since that's the best they can do.

It's also important to realize that there are many different ways to measure EV (all of which are valid). For example, a lot of ppl say that "folding is always 0EV". But that's only true from the perspective of the current decision. If you 3bet 12bb then fold to a 4bet, you've just lost 12bb. From the perspective of the decision the 3b is a sunk cost and folding is 0, but from the perspective of the entire line folding is -12bb. It's valid to measure EV from either perspective. GTO may be choosing between a losing fold and an even more losing call.
Yeodan's GTO / solver output questions and notes Quote
07-21-2021 , 04:39 PM
I replied in the thread you're referencing but I'll say it here, too. GTO is not guaranteed to be >= 0EV in multiplayer games even without rake.
Yeodan's GTO / solver output questions and notes Quote
07-22-2021 , 10:28 AM
Could you give an example of a situation where GTO takes a -EV action? (without rake)

In tombos example,
if the preflop action is +EV
but then there's no +EV action on the flop so we fold.

If we do that infinite times (the same preflop spot, with a different flop) the spot will be +EV not?

Losing once in a spot doesn't mean your play was -EV
Yeodan's GTO / solver output questions and notes Quote
07-22-2021 , 01:36 PM
Quote:
Originally Posted by Yeodan
Could you give an example of a situation where GTO takes a -EV action? (without rake)

In tombos example,
if the preflop action is +EV
but then there's no +EV action on the flop so we fold.

If we do that infinite times (the same preflop spot, with a different flop) the spot will be +EV not?

Losing once in a spot doesn't mean your play was -EV
LJ opens, GTO SB cold calls, BB 3-bets, LJ 4-bets, SB is folding range or close to it, so his cold call is -EV if LJ and BB are going crazy here.

In general the burden of defense against a squeeze falls mostly upon the initial bettor, not the caller, so squeezing too aggressively takes EV away from the caller and gives it to the bettor. There should be many other examples but this seems simple to me.
Yeodan's GTO / solver output questions and notes Quote
07-22-2021 , 02:34 PM
Quote:
Originally Posted by Yeodan
Could you give an example of a situation where GTO takes a -EV action? (without rake)

In tombos example,
if the preflop action is +EV
but then there's no +EV action on the flop so we fold.

If we do that infinite times (the same preflop spot, with a different flop) the spot will be +EV not?

Losing once in a spot doesn't mean your play was -EV
BTN opens, BB calls.

BB will call with hands that make more than losing a blind. This means the postflop EV just needs to be higher than -1bb for calling to be higher EV than folding. The net EV of the BB will be negative.

Now you could argue that the 1bb is a sunk cost so folding should count as 0EV. But as I explained in my last post this is just a shift in perspective. In either case you're looking for hands that have higher EV as a call compared to a fold.

Another example, if you 3bet to 10bb on the BB then face a BTN 4bet, you're in a -EV spot. The hands that 3b-fold just lost 10bb, Even though they took the correct line. Hands that call the 4bet are mostly just trying to recover a portion of that 10bb, most of the calls won't even be net +EV, they will just be slightly better than -10EV (from the perspective of the entire line). Most 4b flatting ranges in general tend to be net -EV overall, yet that is still better than giving up.
Yeodan's GTO / solver output questions and notes Quote
07-23-2021 , 08:31 PM
ur confusing the guy more than needed

Equilibrium will never make a -EV play against an equilibrium player, certain equilibrium plays could be -EV vs non equilibrium players.

Depending on your point of reference, you could say any play that has EV less than the money you already put in the pot is negative, but no one really uses this point of reference and it doesn't actually mean equilibrium is purposefully making a play that loses money, it just means that sometimes your best play will make your final stack after the hand smaller than at the beginning of the hand, like folding at any point after putting money in the pot, but this is obviously sometimes the best you can do, and using a sensible point of reference it's a 0 EV play
Yeodan's GTO / solver output questions and notes Quote
07-24-2021 , 07:34 AM
I think browni's example was pretty good and looks like a spot where a normally +EV open could turn -EV with that kind of action.
If that happens too often the normally +EV open could turn into -EV.
Kinda far fetched, so I doubt this ever actually happens (over the long term)
But his point seems valid to me.
Yeodan's GTO / solver output questions and notes Quote
07-24-2021 , 07:42 AM
Quote:
Originally Posted by Yeodan
I think browni's example was pretty good and looks like a spot where a normally +EV open could turn -EV with that kind of action.
If that happens too often the normally +EV open could turn into -EV.
Kinda far fetched, so I doubt this ever actually happens (over the long term)
But his point seems valid to me.
That's the point, it's super easy to figure out an example in which NE is -EV. It may call a river jam despite the opponent having no bluffs.

GTO is all about figuring out your opponent’s range and developing the best strategy to maximize EV.

NE will NEVER maximize EV in the real world.
Yeodan's GTO / solver output questions and notes Quote
08-06-2021 , 04:24 PM
If I have QTdd on a J97dd board.
There's 50 in the pot, villain bet 50 IP, I have 200 left and villain has me covered.

Is there, pure theoretically, a difference between going all-in on the flop or turn?
Obviously my equity is going to be lower on the turn if I don't hit.
But if I just call the flop, close my eyes and go all-in on the turn, is there a difference between doing this and going all-in on the flop?

If I'm never folding, does it matter when I put my money in?
Not from a strategic standpoint, but a theoretical one.
Yeodan's GTO / solver output questions and notes Quote
08-06-2021 , 08:50 PM
if ur oop you probably just jam flop like always, and yeah it matters because sometimes it can be difficult to realize all your equity oop and you can still make hands with a lot of equity fold vs you when u jam flop. Hands like AT/KQ/KT, etc
Yeodan's GTO / solver output questions and notes Quote
08-06-2021 , 09:08 PM
Yeah I realize I'm missing some fold equity and maybe even value when the draw comes in on the turn and my opponent folds, but that's strategic (or at least how I see it)

I kinda just wonder about the equity realization.
If I never fold the turn, isn't it the same as being all-in on the flop?
This feels wrong, but I can't think of why it doesn't make sense.

Strategically it could be a bad call on the turn, since I'm no longer getting odds.
But if I'm never folding turn, my odds remain the same as on the flop, so why is going all-in there (whether it's a call or shove) correct but then it's no longer correct on the turn.
Even though I end up in the exact same situation.

I understand I half my pot-odds, but the situation at the end of the turn ends up being exactly the same not?
Yeodan's GTO / solver output questions and notes Quote

      
m