Again, I tried to calculate it for one scenario, ofc there is a huge amount of possible scenarios. I like maths and just wanted to know how much of an effect it got. It`s not a lot and maybe if I start learning to code again I might code a programm that does the calculations.

I did look up the cEV of every single hand I might get dealt in the new position with new opponentcallingranges(If I edgefold instead of shoving)In my scenario for example AA/QQ/KK/JJ had these EVs(It`s done quite quick with SNG-wizard):

Hand EV % of being dealt EV x %

AA 3090 0,45249% 13,98
KK 2617 0,45249% 11,84
QQ 2394 0,45249% 10,83
JJ 2199 0,45249% 9,95

Now I wasn`t quite sure how I should handle all scenarios and if it would be correct to sum up the evs of all hands and then multiply it by our bustingequity(In the edgefoldspot before). I might do a post in pokertheory the next days.

It is never chip EV of course.
Additionally we should give us a bit of an edge. But all that makes it not giving up much still.
That is why I think the best play in turbos, in 180-man middle stages for example, in general would be giving up not more than approximately 0.25BB in a hand we push/call when 10BB deep and approximately 0.75BB when around 20BB deep, like I already said earlier.

It doesn't depend on the table that much, as opposed to what you said, because the opponent's ranges, if they are loose or tight, are counted in already. If there are looser players, smaller stacks or our stack is bigger we adjust accordingly and still get those numbers for how many BB's to give up.

So it actually can be generalised, in general. :P
Anything further is a bit too complicated to talk about atm.

This is part of what I was getting at. If you can calculate the probability you will be dealt a hand in a better +CeV spot within the next orbit (before the blinds hit you again) you could work out whether folding a +CeV edge is +$Ev (and I think a lot of the time when the edge is small it is). Although I'm not sure how to do this since I'm not great at the math and it would seem to get a lot more complicated if you factor in the effect going through the blinds has since you lose fold equity and have a lower chipstack so can't double up as much.

Yeah, it`s a lot of factors. It`s way more intersting with ICM though because then mistakes of opponents need to be counted too.
I think on Sunday I might make a bigger post in Poker Theory, there are a lot of guys that are much better in maths than me.

First, I would like to thank TakeMultiBrah for starting such an interesting thread and all other (like Nisas, Maniac and all others) for contribution.

If I understand you correctly then:
You can't calculate how much likely it is to be dealt better hand in next orbit. According to Theory of Probability(discrete here as sample space is finite), every round is mutually exclusive and independent. So, probability of getting dealt AA is out of 52 cards is the same in a particular round as it was in the last round. Not, getting dealt AA for 2 orbits does not mean it'll increase the probability of getting dealt in next orbit. So, if you decide to shove top 20% of your hand at some point as you are so short, mathematically you can't expect that, you'll be dealt top 20% at least in 1 round out of 5 round. You can be dealt top 20% of the hand may 5 times in a row or may be 0 times in a row as the distribution is random. Each round is independent so they won't influence each other.

By better +cEV spot, if you mean position, opponents and having way better edge in terms of showdown equity(and fold equity when you are shoving) over their calling/shoving range, which you might not have against some other solid reg's range, that's different.

It doesn`t matter, it`s like "I lost 10 80/20 allins now I have to win", does it really matter? No. We only try to make +evdecisions.

Longterm we will(or at least should ) get it dealt 20% of the time, it doesn`t matter if we don`t get it dealt in a single round.

This is not true either.

For the new Hand you just need:

- Shovingranges for your opponents
- Callingranges for your opponents
- The new stacksizes

If you got these things, you can calculate the theoretical ev of every Hand. In my Example if everyone folds and we can shove JJ+ got these equities:

Hand EV % of being dealt EV x %

AA 3090 0,45249% 13,98
KK 2617 0,45249% 11,84
QQ 2394 0,45249% 10,83
JJ 2199 0,45249% 9,95

That`s only the scenario that everyone folds, if someone shoves in front of us then AA got a different EV.

It`s really late here and though to explain in english for me.

your most of the argument is very nice and beneficial for me, but not all.

This is understandable and justified.

When you see a very large sample size, you might see you got 10% of the hand ~10% of the time. But, when you are calculating expectation of getting dealt top 10% of hands in 1 out of next 10 hands in next few orbits, the sample size is really too small. if you can survive next 10 million orbits, then you can expect to be dealt top 10% hands ~ 1 million times.

Same argument as you are saying before that 20K tourny is not large enough sample size to decide a player's win rate, which is actually correct.

So, my argument was when you are doing mathematical analyse of different spots to make more +EV decision, taking into consideration the future possibilities of dealt cards and others in that tournament, we should follow the theory of probability correctly. This is not my thought, this is Theory of Probability and I can bet on that.

The sample size doesn't matter in expecting what happens, it is important for statistics only. (Sorry, I may not be good explaining things in English.) For the smaller sample size the only difference is the variance, that is bigger. But the variance is not important in what is profitable.
We should always act based on what is expected on average.
If you're holding 72o in your hand we all know the next cards you will get are better ones on average, and if you are getting like 32o - 94o cards the whole orbit long we all know that the next orbit you are getting better hands on average.
The next cards/orbit we are getting is not related to what we got now and doesn't HAVE to be better nor worse, it is exclusive and independent to the current cards/orbit, as you said correctly, but since the next cards are the average we can get and since we got 72o at the moment, the average, we are getting next time, is better than the below-average-72o we got now.

That said, often enough we can't wait for better hands nevertheless.

I agree with you that you usually get better hand on average, but I think my point was different. How will you fit this expectation in calculating +EV spot??

I am trying explain my point by example:
Suppose we'll shove with 80% from SB, 60% from Button, 30% from CO and 21% from MP1 in particular spot as we calculated it is profitable regarding our stack and opponent's stack and their calling ranges. Now we didn't get dealt top 21% of hand in MP1 and we folded. Will that increase our chances to get dealt 30% hand in CO?? Mathematically (Rather according to probability theory), No: you can't predict, the chances of getting dealt top 30% hand is 30%. It won't go up as you did not get top 21% hand in last hand. So you might expect to get better hands next time but you can't put it in a formula.

I think I am clear enough now. Please let me know your opinion(with mathematics and logic) if you think I am wrong.

Yes, it is absolutely correct what you said about those ranges and how it doesn't affect cards being dealt next hand. I think the vast majority of us knows that but this was well explained, thank you.
But I think we can put the average we can expect from the next hand in to the formula. We are not expecting some mysterious better hand we can't calculate, we are always expecting the average.
The formula would be just too long and complicated if we would try to find the formula of winning the whole tournament or even then if we'd like to do better just in the next couple of orbits. Because there are too many factors to be considered. We'd have to know the opponents that well we never could to predict their actions and even then the formula would get too complicated. At least for me. :P

I think the ranges we are normally using in our game are naturally close to those we could calculate in those formulae anyway. At least it is that for the regs. Because we derive those ranges from our experience of winning and so on. Although the variance could give us the wrong ideas about good ranges.

I haven't read much books about this, (don't read much books tbh. Just using my logic, sry. So please take this as my opinion only).
I am sure it could have been calculated roughly in some books I am sure.
And I am sure some of the guys here can calculate the rough estimates too for a few consecutive hands/orbits.
It is just complicated and would take time.

I think the basic ranges we are pushing/calling and everyone knows about are the result of the rough calculations someone has done and the natural adaption anyway.



Gl at the tables everone!

/thread?

Nice post

However you can calculate the probability of being dealt a better hand in the next orbit and each round is not independent from the players' point of view eg if you fold the sb everyone can see that you are not pushing atc and this gives you more fold equity on the next orbit.

PS The chances of not getting a 20% hand is obv .80 so the chances of not getting one in the next three hands is .8 x .8 x .8 ie 51.2% so you can expect one roughly every three hands.

This is correct, I agree with you, but not contradicting my Mathematical logic at all.

Sorry man, but this is completely wrong. you did it in the way we calculate the fold equity if 3 players acting behind us and have 20% calling range over our shove. You can do that there as in those cases cards dealt to them are dependent on each other( e.g., if two people got AA then the third one can't get any A in his hand). It would be easier to explain the concept of 'independent', 'mutually exclusive' through Venn diagram. Then it will be clear. But may be this is wrong thread to discuss it. Poker theory would be right one.

Draw all the diagrams you like but it's still right. Let's make it simpler. If you chose 50% (like a coin toss) instead of 20% there is a 75% chance of getting it over two hands (.5 + (.5 of .5)). I think you will agree that there is a 75% chance of getting tails at least once over two coin tosses.

It's an MTT part of the game is knowing 99% of the time you're gonna cram some middle range hand and have to win a flip. You can't fold cause you get knocked out a decent $ of the time :/ Also more than 1bb cEV is HUGE.

Read Kamel's post he is smarty

Quoting/bumping it for the soul¡!¡!¡

Me too would like to quote kamel again.

The biggest and the most common mistakes, in terms of the overall $EV, most players are usually making lie in not finding all the +cEV situations (hard to read the opponents' ranges, hard to adapt, adjust and calculate your own), and are not lying in not giving these situations up often enough when we find these. (Sorry, too many the "not" words.)
Although we should risk less when we have a big edge we can't lose much in a zero or sometimes a slightly -cEV situation mid stages but we can lose more if we miss a lot of the +cEV situations constantly.

So we should be more worried about finding the +cEV spots and reading the opponents more precisely, as accurately as we can, than passing these spots, as long as these are not -cEV.

All this is obvious to everyone though.

Gl.

"Just to give some classic examples. To give up the race QQ vs AK, you would need ~250% ROI (from Mathematics of Poker IIRC)."

That doesn't make sense taken out of context eg if it's the first hand of the PS Sunday Million folding will cost about $12 in equity. I'll be a monkey's uncle if you can get 250% ROI from that.

bumping, this thread need some more love :P

Amen!

When its part of your nash equilibrium shove strategy?



I had a buddy who was a really good basketball player where we come from. Told me a story about how he went to his coach to say he watched Micheal Jordon give advice on aiming at the back of the rim to make a shot. But his coach explained that if you keep shooting for the edge of the basket, eventually you'll start to hit it, and the ball with just start bouncing off the hoop. He suggested instead my friend should aim to shoot through the basket. Took me many years to understand why I should take the coach's advice over MJ (was this a cool story?).

Edge is not what the OP thinks it is. If we have the perfect answer that we could apply a certain x% adjustment to, that would be great. But we don't have the perfect answer in the first place. So to apply an arbitrary edge is arbitrary.

One player might see the field as too loose but apply a -x% edge
Another player might see the field as too tight but apply a +x% edge

Both come up with the same answer of what hands to shove.

Since there are too many variables and we are all human, looking for a certain % edge to adjust too is really just adding voodoo in your game.

The quickest way to making optimal decisions is to learn the fundamentals lines from a good player, and build your math behind it, rather than the other way around (OP).

The longest route would be to deconstruct the myths on the game, and construct your play based on what the game really is.

lot of good info here , going slightly off the original topic although it has been touched on in this thread. i am interested to hear more opinions on possible rois in the current climate. the 2 'best' players mentioned have ridiculously good stats in a host of games (one has made 200k+ this year) but in 2013 they have played less than 23k 180s between them (20k and 3k) and the stats will be skewed by 3r. please note i am not saying that these 2 would not be the best at 180s, it just seems they dont play them anymore so its hard to ascertain possible 2013 rois from these guys.

if anyone would like to hazard a guess at possible rois for the following it would be much appreciated:

$3r
$8
$15
$35

You shouldnt pass $EV+ spots with the argument that a better one is coming in the near future.You know you can take both if you win the first one and fire up another trny if you loose where those same spots u are waiting for will come up with the same frequency"on avarage".Poker you know is one long,long(i.e discontinuous) game as long as you are at the tables.

in all honesty i could not disgaree more. why take a small $EV advantage if you have a massive skill advantage. could rhyme out 1000 scenarios where it best to let it go.