At the start of the tournament ICM is pretty irrelevant afaik.

The important factors are the edge you have over the field and how fast the structure is, the faster the structure the more edge you can have with bigger stack due to getting in to the +ICM spots.

Assuming we have good edge, but structure is fast, we might choose to flip to gain big stack which we can use to roll over the table.


I don't play MTT's as I feel like killing myself everytime I reg for one, so this is just my guessing lol.

I mean no offense, I actually thought you completely skipped that part.

The "wait for a better spot" is pretty easy to demonstrate. Let's say you can have a 1% edge in this hand or bust but now we have the information that a guy is going to be open jamming every hand and you will get a 75% equity spot in the next 10 hands on average. You will add 50% of your stack by this hand but you need to be in the tournament to take advantage of it.

So what happens if you take the flip and bust? Well you're going to have 200% of your stack 51% of the time and 0 49% of the time, if you are still in you can take advantage of the 75%. We're looking at chip EV only.

Your EV (0,51 * 200) - 100 + 0,51 * ((0,75 * 200) - 100) = 2 + 0,51 * 50 = 27.

When you don't take the flip your EV is simply 1 * ((0,75 * 200) - 100) = 50 because you get to that spot every single time while when you take the flip you can only take 50% advantage of that "better spot".

Of course you don't know someone is going to open jam but if you know for a fact you will be able to take advantage of weaker players over and over and over then it is very well worth it not to bust.

OK, wow, this is exactly what I needed to see to convince me about the "wait for a better spot"! It really is hard to argue with the math and you've left me with no choice but to agree with the "wait for a better spot" strategy.. So this basically also means that it is a terrible play to call this all in in a cash game? Since you know your opponent is ******ed.

It seems that some of the people in this thread were also wrong as I was so I hope they have learnt something new as well and that this thread wasn't just a huge time waist.

Sorry for all the arguing. Please lock this thread up.

NOOOO, I was wrong again. You can rebuy in a cash game so "waiting for a better spot" doesn't make sense in a cash game?

Also we can solve how much of a favorite you need to be to take it instead of wait for the 75% hand where we have an EV of 50.

EV = 50 = (x * 200) - 100 + x * 50

x * 200 + x * 50 = 150

250 * x = 150

x = 150/250 = 3/5 = 60%

So there we have it, you need to have 60% or better to win that hand in order to risk missing the 75% hand.

Not in strict theory. If it keeps a fish at the table, and there aren't an endless supply of fish, then it could be correct if it's a very very small edge.

Yea but I'm talking about a purely hypothetical case when the fish will continue playing like that forever and everyone has infinite bankrolls. Also there is no rake. I just tried to make things as simple as possible. And I understand that this is far from reality..

Actually this is the first time I've ever done any calculations on this but the result is quite there. We can substitute the 75% hand that gives you an extra 50% of your stack on average by saying "this player has 50% return on investment". ROI is how success in MTTs is measured and of course it isn't decided in one hand but if we're flipping for the stack in hand 1 of a tournament, half the time you aren't going to achieve your ROI which is actually sick.

So this is of course at the start of the tournament when there is 8 days of play left potentially. At any later stage the time for you to achieve that edge gets less and so the ROI from that point on becomes less. Let's say half the players are eliminated you will have less than 50% left for your ROI, let's say 25% as an approximation. For you to miss out on a 25% ROI you can take all ins that are more close to a flip.

The math works out to be 125/225 = 55,5%. So the further you progress in the tournament the less you should avoid flips. It makes sense but it's the first time I've actually tried to quantify it.

Edit: a 10% ROI player can take a 52,3% edge and a 1% ROI player can go as low as 50,24%.

This ignores the fact that later in the tournament ICM tends to be a bigger factor, but for e.g. winner takes all tournaments it's a really useful model.

Cool, this is a nice calculation, I like it . However, this "wait for a better" spot strategy only maximizes the winning of a single tournament. You can also go play another tournament if you are eliminated similarly to rebuying in a cash game.

But yeah, this means that you should try to wait for a good spot versus a fish, because there might not be a fish in the next tournament to exploit.

Yes, it is maximizing the ROI for that specific tournament. If you look at the hourly rate you might as well register another tournament. Also ICM does make a bigger impact as you get closer to the money but in early stages it is probably shadowed by what I just calculated.

OK, so here we should basically ask ourselves this:

DOES INCREASING MY STACK FOR A FACTOR OF 2 INCREASE MY ROI FOR A FACTOR OF 2 OR MORE? If the answer is yes, you should call the all in. If the answer is no, you should fold. Simple. The hard part is determining the ROI of the double stack I guess

If somehow the ROI of a double stack is more than 2x your initial ROI than we should absolutely call the 33 vs AJ all-in..

Did you consider that having a 2x stack in the second hand of a turney will increase your ROI? It probably isn't 2x the initial ROI though, as discussed previously.

Of course, getting extra chips will definitely increase your ROI. The problem is that the increase is probably not as valuable as the loss.

Funny sidenote, a losing player with a -20% ROI has to snap it off like a boss when he has 45% equity or more.

Yeah that's funny . In theory -20% ROI players shouldn't be playing poker at all.. Anyway, doesn't your (100+ROI)/(200+ROI) calculation only maximize the winnings of just a single tournament? In reality you can be a 50% ROI player and snap-call a 55% flip in the first hand and just go play another 50% ROI tournament if you lose the flip?

I think the real formula for this problem is:

if ( x*ROI_{double} > ROI_{initial-1BB} ) call;
else fold;

Bolded part is the problem of your calculation. You assumed that having a 2x larger stack will result in the same ROI (50%) in this case. The calculation was OK for the case when a player goes all-in every hand, but I don't think it translates to ROI very well..

This is why it's a call in a $10 online tournament (assuming you're properly rolled) but a fold in the WSOP ME.

EV dominates RoR in online play where you're correctly rolled, but in a case like the ME, your immediate RoR dominates small EV edges.

DUCY?

Hmm, it might still be a bad call in a $10 online turney, because the ROI of your double stack is probably not 2x the ROI of your initial stack.

I said RoR not ROI

Risk of Ruin

Yeah, I know but calling all-in with 51% on the first hand is probably a bad play regardless since your 2x stack will not increase your average profit by a factor of 2.

Fair enough, but it's a much closer decision if you're going to play 2,000+ similar events.

That's looking at one hand. You're in a tournament, not a poker hand.

Your question is overly simplistic. If I said you could have 1 dollar right now, or 100 dollars in 5 minutes, which would you take? Your logic might be "How can getting a free dollar ever be bad?" It's not bad in a vacuum. It's bad in context because you gave up something better.

Now your hypothetical question is fantasy because no one is ever going to show you their hand like that. But to compare apples to apples, if you knew a situation like that was going to happen every single orbit, then would you take it now? If you knew that every orbit, it was going to get folded to the SB, he was going to show you his cards and then shove, would you take the AJ vs. 33 hand?

You should not. The risk of busting out is too high for such a tiny reward when there is a much better bet waiting for you down the road.

That's basically how pros look at these situations. They would much rather play 100 hands for 5% of their stack each time when they're 60% favorites, than to play one single hand when they're a 52% favorite. It's very difficult to lose in the first scenario, but pretty easy to lose in the second.

If you have unlimited bankroll and don't care about the variance, then yeah obviously this is a good call.