Response to Skodljivec latest video

Hi,

great post! My plo mind always goes in those spots wether to call pre, or stackoff. Like you said, it's very interesting that a shove is more + ev than a call. It's actually something that never crossed my mind, as plo players are tought to call 4b or fold, not considering the wierd spots where a shove is actually better on EV than a call.

But i should probably fold there preflop. I think that's the best play in the long run. I'd probably muck that at plo100, but i called at plo25 if that makes sense ♦.

Regards,

Skodljivec

Make sense, makes the videos a little bit more interesting

What is the trashiest hand you would go :
"aaaaw, shiiiiiiet. I have a good hand, you're short. Ship!"?

One other sick thing..
Say we are up against random aces, a flopcall can generate less loss of money

We have 12,5% against random aces.

Ev : 0,125*122bb-0.875*24bb = -5.75bb

Really nice post!!

That said, it's better to go broke PF vs 4bets from shorts than calling? But what's the "deadline" in terms of BBs? If he had 70BB what's the best play? Still shoving? And vs 80BB stack?

In this scenario, I think it is.

Above 70bb, I think we should play as the hand is played.

You gotta have a mistake somewhere.

Cos villain is never folding on flop. So he will shove 100% of flops. Now if we call pre, and stack off on every flop, the EV is same as shoving preflop.

But if we can fold at least one flop, it's better to call than shove preflop.

I will look deeper into this in a bit.

(Something funny with the hand history, as the stack sizes doesnt add up to actions)

Ok, first thing. Folding when facing 4bet has EV0.

To stack off preflop, we would need little over 40% equity (rake isn't counted). As we have 39% that is -EV. So you have a mistake there.

We would need to invest $12.15 to shove, to win a pot of $30.25.

If we call, at flop we have a situation where the pot is $18.60 and we have $6 behind. So to call villain shove we need 20% equity or more. We get that around 65% of the times (I rounded down, so actual EV is little higher) and as Oracle says we have around 53% equity.

So 35% of the times we fold flop and lose that preflop call, which is -$6.20.
65% of the times we go all-in on the flop, and we invest $6 (plus preflop call $6.20) to win 53% share of $30.25 which is $16. So our EV is +$3.80.

(35% x -$6.20) + (65% x +$3.80) = (-$2.17) + (+$2.47) = +$0.30.

So as shove pre is slightly -EV, calling is slightly +EV. If we deduct rake (which is around $1.35) even with supernova status it's better to fold pre than call/shove, IF we only think EV for this hand. Metagame might change this.

If we are 100% sure he has AAxx range, there isn't much difference between calling or shoving. If there is at least slight chance he has something else, shove pre is better.

Hi,

i have to admit i didn't look at the calculations closely cause i was in a middle of a session when i was replying (great role model right? ♦).

But it does seem ilogical that a shove would be better than a call tough. I'm not much for math, that's something people know about me, but i do know b heart the preflop equities of hands with simmilar structures preflop vs AAxx, so i knew it would be close when i was playing.

Still i applaude the effort, and i'm gonna leave the % calculations to you mathematicians. I'm gonna go crush some fish in the mean time ♦

Regards,

Skodljivec

correct equation should be:

EV = (1 - top_x)(-pf_bb) + top_x{av_equity(totpott) - (1-av_equity)loss}

Which results to:

pf_bb = 24,8
top_x = 0,64
av_equity = 0,53
totpott = 122
loss = 24,8bb

EV = (1-0,64)(-24,8) + 0,64{0,53*122 - (1-0,53)*24,8}
EV = 0,36*(-24,8) + 0,64*(0,53*122 - 0,47*24,8)
EV = -8,928 + 0,64*(64,66 - 11,656)
EV = -8,928 + 0,64*53
EV = -8,928 + 33,92
EV = 25bb

Preflop allin should be ok.

So its even more profitable to call the 4bet, and thats what most would expect here.

Did you remember to deduct that flop all-in, $6 too? That doesn't get into the pot itsef nor it's counted towards our winnings. Preflop allin is -EV, period. Not much, but it is. Some simulations might give us 40% equity which makes it barely +EV.

I think you forgot to take that $6 away from the winnings when hero hits the flop. Which would lead to +1bb EV, just like mine did.

You are right. I cant win the total pot inclusive my call. So its totpott-loss (122-24,8).

So it should be:

EV = (1 - top_x)(-pf_bb) + top_x{av_equity(totpott-loss) - (1-av_equity)loss}

Which results to:

pf_bb = 24,8
top_x = 0,64
av_equity = 0,53
totpott = 122
loss = 24,8bb

EV = (1-0,64)(-24,8) + 0,64{0,53*(122-24,8) - (1-0,53)*24,8}
EV = 0,36*(-24,8) + 0,64*(0,53*97,2 - 0,47*24,8)
EV = -8,928 + 0,64*(51,516 - 11,656)
EV = -8,928 + 0,64*39,86
EV = -8,928 + 25,51
EV = 16,5bb

But its still more than your +1bb EV. How comes?

You take that $6 (or 24bb) away from your expected winnings. You are not investing just portion of it, you are investing 6 dollars to win a share of pot+villains money.

Think it this way:

We are investing preflop and flop call/bet. Flop call goes into pot anyway, but on the flop we have the decision to put the rest or not. So at that point, the $6.20 is already in the pot. We are risking 6 more dollars to win a 53% share of total pot. Our expected winnings is $16. And as in any investment/winnings situation, to get the total value you take investments off from the winnings. As far as I understand you have deducted only portion of your flop investment.

I suck with formulas, that's why I like to use common sense examples to check my thinking process.

Let's say we play with dice. We both put one dollar to a pot. Then we roll the dice. If 1 or 2 comes, you win the pot. 3-6 comes, game continues.

If game continues, and this time we put another dollar to the pot and roll the dice. 1-3 I win, 4-6 you win for the whole pot.

What is my EV? I can't put that to your formula, but you can check if it's right.

Correct answer should be -1/3 (or -$0.33). If your formula gives something else, there is a flaw.

The handhistory is from the video, therefore there is some misstakes.
I also see that i put the wrong amount of preflop bb.

I don't think there is anything wrong with EV-calculations (except wrong pr_bb, the EV is then more like -5,5 bb)

The major mistake I did is that I calculated the EV IF we know that villian has AAxx preflop and not from the point on when it dawned on us that he had AAxx and our preflop investment already belong to the pot.

All in all.
- If we know that Villian has AAxx it's a losing play (impossible to know unless cards are faceup)
- From the point on when we realise that we are up against aces, the most profitable play is as played (proven by kyyberi)

We should always think our decisions based on the information we had at the moment. For these kind of hands, we start the calculation from the point where opponent has 4bet us. The money that is put into the pot prior to that isn't ours anymore.

These kind of calculations are the very core of winning poker in PLO. I don't know for sure about the higher limits, but at PLO25 and everything under that the biggest single leak (even for most regs) is to overplay their aces. And that overplay is just an outcome of not understanding the mathematic realities behind it.

This thread is old somehow, but I made some work to present the ways to work through this with odds_oracle and PQL:

https://forums.pokerstrategy.com/de/forum/thread.php?postid=13791660#post13791660

Some text is in german, but youll get the point there. Maybe its usefull for some people, especially regarding odds_oracle and PQL.

Nice work Kreatief!

Do you know the PQL formula to give the avg equity on the flop?

yep. Its this one:

So if you only want to know the average equity on flop (not only when at least 19.6% equity), then its:

select avg(HvPerceivedRangeEquity(hero, flop, "AA")) from game='omahahi', hero='KhJsTd9h'

If you want to add the constraint with the 19.6%, then use the where clause as well:

select avg(HvPerceivedRangeEquity(hero, flop, "AA")) from game='omahahi', hero='KhJsTd9h' where HvPerceivedRangeEquity(hero, flop, "AA")>=0.196

But as written, this can also be done in odds_oracle without PQL, by using the require-constraint at the top of odds_oracles main window: