Hi,

I still do not get why the BF is so low at 4 and 3 players left. The payjump is so much higher 4-3 and 3-2 then it is for 9-8. Therefore icm-wise we need to call allins a lot tighter when 3 and 4 handed then 9 handed. So i think the bubble factor is also much higher in these cases then it is playing 9 handed.

I did some calculations using hrc and i dont come to your numbers (for the 180s).

for 9 man left i get: 1.58 and for 3 left i get 1.71.

Dont know if anyone follows this thread. but i was just wondering.

No offense but this is such a useless thread. No one in this sub-forum is going to be doing this in real time. Average bubble factor is a useless measure when everyone's stack is going to be different; therefore, value differently so bubble factor is always changing.

Those calcs are closer to what you get if you forget to give the person who busts out his prize - though still not the same. Could you post your workings?

3-handed is easy to check using the method in post 20.

AvBF = 1+((2nd-3rd)/(1st-3rd))

ICM isn't just about the next step specifically, it's also a measure of the extent to which the remaining (i.e. not already locked up) money doesn't just go to first (and then the extent to which the money not going to first doesn't just go to second etc.).

So heads up, 10% of the original prize is still available (50% has been paid out but the people left have also locked up 20% each so are just fighting over the last 10%) - all of that 10% goes to the winner.

3-handed, 27.2% of the original prize fund is still being fought over, of which 68.4% goes to first and 31.6% to second.

Prize: Money Remaining: Share Of first

30:
20: 10: 100
11.4: 27.2: 68.38
7.4: 39.2: 57.65
5.8: 45.6: 53.07
4.3: 53.1: 48.4
3: 60.9: 44.33
2.2: 66.5: 41.80
1.5: 72.1: 39.53

So the increasing "share to first" as the FT progresses counterbalances the growing "next jump(s)", and is why 180 mans have a fairly even bubble factor all along the final table.

It's not just next-jump and money to first - the whole shape is important.

Also SB vs BB 9 handed with even stacks 7.5BBs deep, antes 0.1BB

http://www.holdemresources.net/h/web...&s9=30000&s10=

SB 86.4%, 22+ Jx+ T2s+ T4o+ 92s+ 95o+ 82s+ 84o+ 72s+ 74o+ 62s+ 64o+ 52s+ 53o+ 42s+ 32s
BB 36.5%, 33+ Ax+ K3s+ K5o+ Q7s+ Q9o+ J8s+ JTo T9s

and 3 handed
http://www.holdemresources.net/h/web...=&s8=&s9=&s10=

SB 80.1%, 22+ Kx+ Q2s+ Q3o+ J2s+ J4o+ T2s+ T5o+ 92s+ 95o+ 82s+ 85o+ 72s+ 75o+ 62s+ 64o+ 52s+ 54o 42s+ 32s
BB 30.6%, 44+ Ax+ K5s+ K7o+ Q8s+ QTo+ JTs

or if we treble the antes to take account for our being short-handed

http://www.holdemresources.net/h/web...=&s8=&s9=&s10=

SB 84.6%, 22+ Qx+ J2s+ J3o+ T2s+ T4o+ 92s+ 95o+ 82s+ 85o+ 72s+ 74o+ 62s+ 64o+ 52s+ 53o+ 42s+ 32s
BB 38%, 33+ Ax+ K2s+ K5o+ Q6s+ Q8o+ J8s+ JTo T9s

So the difference is not that major, as is consistent with the OP.

If we do the same thing with the 45s payouts we get:

9-handed
SB 100.0%, Any two
BB 27.6%, 44+ A2s+ A4o+ K6s+ K8o+ Q8s+ QTo+ JTs

3-handed

SB 82.8%, 22+ Jx+ T2s+ T5o+ 92s+ 95o+ 82s+ 85o+ 72s+ 75o+ 62s+ 64o+ 52s+ 54o 42s+ 32s
BB 43.3%, 22+ Ax+ K2s+ K3o+ Q4s+ Q7o+ J7s+ J9o+ T8s+

so there is a major difference. Note also that only 7 cash in a 45 man, so even though the next pay jump is zero, it's the more triangular shape of the payouts left to come that is crucial.

None taken.

TBH I have seen people over-use these numbers in the MTTSNG sub-forum, using the AvBF when they are 2nd in chips clashing with the chip leader. The real bubble factor is the right one but as you say, you can't do it in real time.

The average numbers which you can do before the tournament starts are good to give you a quick overview of how heavy the different stages of the tournament are - and that is one of the things that plugs into your subjective on the spot judgement of the ICM dynamics of the actual stacks at the table.

It's an okay indicator. An ICM calculator would be better as you can adjust ranges on the fly and really examine the way table dynamics work in different stages of an sng or tournament.

in HRC ( not the web version, but the paid product)

4 man.
180 payout structure
equal stacks

Eq_current: 19.87
Eq_loose: 7.4 (forth place)
Eq_win: 27.23


BF = (Eq_current - Eq_loose) / (Eq_win - Eq_current)

BF = 19.87 -7.4 / 27.23 - 19.87 = 1.69.

You have 7.4 as 4th place so I assume you're going off the percentages I quoted then.

Payouts 30, 20, 11.4, 7.4

Equity 4-handed = 30+20+11.4+7.4 / 4 = 17.2

Eq win - we have 50% of the chips, so 50% we finish first, 33% we finish 2nd and 17% we finish 3rd.

=(0.5*30)+((0.5*0.667)*20)+((0.5*0.333*1)*11.4) = 15 + 6.667 + 1.9. = 23.5667

BF = (17.2-7.4) / (23.5667 - 17.2)
= 9.8 / 6.3667 = 1.54

If HRC is giving you incorrect answers then you should take it up with them

hrc is not incorrect. your math seems off. i will respond later.

i dont have time now to respond in depth. but it is obv that your numbers are wrong. if the bf is about the same 9 and 4 handed that means that we can allins with about the same range. This is clearly not correct. Because the payjump 4-way is much higher and thus we need to call tighter then 9way.

These are the numbers i get from a 4 man 180 payout equal stack:
Where EQ is the equity in the TOTAL prize pool. (not the remaining, and i think that is were your mistake lies)

EQ_current = 19.87
EQ_loose = 7.4
EQ_win = 27.23

Bf = EQ_current - EQ_loose / EQ_win - EQ_curr

Bf = ( 19.87 - 7.4 ) / ( 27.23 - 19.87 ) = 1.69

When i calculate this for 9left i get a bf of 1.58. Those numbers seem pretty right to me. But please tell me if you think they are incorrect.

The biggest pay jump of all is from 2nd to 1st, so presumably that's also the biggest ICM pressure then ...

As I write earlier, ICM bubbles factors are firstly a measure of the extent to which the money still being fought over doesn't just go to first place. The shorter handed you are, the more of of it goes to first. 2-handed is a good example of which is the strongest force.

The remaining prizes total 68.8, the players don't have EQ of 19.7 each because then the total prizes would be 78.8 (exactly 10 more, perhaps you have a typo in the payouts here?)

Independently from me, ICMizer online gets the same result BTW:

http://www.icmpoker.com/icmcalculator/#Eufc

Again independently, in this old thread they give the BF for the final ten in the old payout structure (they used to pay 18 people only) and the new ones paying 27 people and the results are consistent with mine:

http://www.internettexasholdem.com/p...tor-65239.html

Either HR calculator is wrong or you are using it wrongly.

Thank you so much, I've been looking everywhere for this!

If our calling ranges got tighter every time the pay jump got bigger, our HU game would be pretty bad.

obv there is NO ICM pressure HU. So the bf is 1.0 in that case.

Again, the bf determines your real tournament odds. Say we are 4 way 10bb deep in the bb. The b shoves. Now our chip ev odds are 1.3 : 1 (we need 43% win % )
Our real tournament odds are 1.3/bf : 1. If our bf is 1.5. then our win chances need to be: 54%. So our bubble factor directly affects the win % needed (and thus our calling range).

So, if you say that the bf_9left and bf_4 left are about the same. You say that the range we need to call the b shove is about the same. This just is not correct. Its horrible actually. We need to call tighter 4way then 9way. Try it out, use hrc or icmizer. And see for your self. Thats because the bf is larger 4way then 9way.

Thank you so much for this

I play these and i dont fully get why neither.

I dont know if the tiop heavy structure of these 180s have anything to do with it where almost 82% of the prizepool is concentrated in the top 7 paid spots and 61.4% in the top 3 def something i would like to know more in depth, cheers.

Im just getting the hanf of it but as i read Leia is asuming equal stacks and lektor is asuming we have 50% of the chips in play 4 handed.

No, those are two different lines.

Equity 4-handed (17.2) is when we have all have 1/4 of the chips and therefore equity.

Eq win means equity if we win a flip and therefore have 50% of the chips in play 3-handed. Our equity is then 50% of 1st prize, 33.3% of 2nd and 16.7% of 3rd.

The calculation for a Bubble Factor is

(initial equity - equity if lose) / (equity if win - initial equity)

So when calculating the average bubble factor - i.e. when everyone has the same stack and we are 4 handed

initial equity = top 4 prizes divided by 4.
equity if lose = 4th place prize
equity if win = our equity in the situation where we win a flip 4-handed and go to being chip leader 3-handed with the chips 50-25-25.

I'm not sure Leia is doing any calculation. He's just reading off the equities from a piece of software that's set up wrong. 4-handed with equal stacks we don't have 19.87 equity each when the total prizes we are playing for add up to 68.8.

This is completely correct. /thread

no it is not.

If it were then the ICM pressure at 7handed would be bigger then when we are 3handed. But if you fire up HRC and check you will see that vs a fixed shoving range we have to call tighter 3 handed then 7 handed facing a shove (you have to adjust the ante somewhat to get the same preflop pot size).

How do you explain that?

he's a she. and it is not setup wrong. your numbers are wrong. numbers are nice, but they need to make sense. your numbers are saying that we need to call off shoves tighter 7way vs 3way (if you only look at icm). Just try out some spots, and you will see that that is not the case. the reverse actually. So how can that be?

Everyone has 10bbs 7 handed:



Everyone has 10bbs 3 handed:



As the BB vs a SB shove I can call wider 3 handed because there is slightly less ICM. Standard 180 man payout structure. I'm sorry, but you're not correct.

My apologies.

If there is 68.8 left to pay out and everyone has the same stack, (ignoring position) the only number that makes sense for everyone's equity is 17.2.

I assume comparing "like with like" is going to depend on what you do about antes. If in your calculation there are seven lots of antes rather than only 4 it might explain why your results are different to Sandman's - but that means we have a lower effective stack relative to the CPR (total blinds and antes) and it isn't the same thing as bubble factor.

It's correct, but of course only within the constraints that ICM itself is accurate.

4way:



7way:



So 4 way:

BU: calls 13% tighter
SB: calls 11% tighter
BB: calls 14% tighter

i agree with you here. i artificially upped the ante abit to create the same size preflop pot as 7way. Otherwise you cant compare the two cause the bb just gets a better price 7way vs 4way. But we can also leave out the ante altogether. Actually this is a much better approach to compare the two ICM wise. Then we get the following call ranges for a 25% co Shove:

7way:

B:5.0%
SB:5.5%
BB: 7.1%

4way:

B: 3.8%
SB: 5.4%
BB: 7.1%

Now the BB call range seem the same. But if you at how negative certain combos are you can see that there is still more icm pressure 4way then 7way. For instance: A2o:

4way: -2.7
7way: -2.4

Your screenshots seem to be comparing a shove from the Hijack 7-handed with a shove from the Cutoff 4-handed.