Thoughts on statistics

[Scuba Chuck]

I'm noticing some interesting consistencies in my results over the past 2 months, and first 10 days of this month. Pretty consistently I'm in the top 50% for my top 4 finishes.

February Results:
45% ITM, and 9% 4th place finishes (308 games)

March Results:
39% ITM, and 12% 4th place finishes (475 games)

April (Month to date) Results:
35% ITM, and 15% 4th place finishes (355 games)

So, here's my question. Do most of you place in the top 4 in 50% of your games? Any multi-tablers out there hitting high 50s or 60s?

[spentrent]

I'm in the top 4 almost exactly 50% since I've started playing profitably. It seems like 5% of those are sometimes 4, sometimes 1, and that seems to separate the good weeks from the not so good weeks.

Just guessing about the nature of that 5%. (~200 games per week if that means anything.)

EDIT: 4-tabling. (200 a week otherwise would be insane /images/graemlins/smile.gif )

[johnny005]

after 550 20+2's I'm 49.5% for top 4 finishes, and thats after a bad streak of 1%ROI after 230 games..

[Freudian]

40% ITM
18% 4th

I have great expertise in going out in a blaze of glory in 4th.

[johnny005]

I'm not sure what is better to haev a higher % of 5th or 4th? Anyone who has played a high # of these what % are they getting for 4ths?

[Freudian]

I have 17% 5th places also. So top five 75%. Which probably means I am a bit too conservative levels 1-3.

Anyway, my sample is only ~600 SnG's (new computer, new db). So I suspect the numbers are far from set in stone.

[spentrent]

[ QUOTE ]
I'm not sure what is better to haev a higher % of 5th or 4th? Anyone who has played a high # of these what % are they getting for 4ths?

[/ QUOTE ]

Any amount is too many. /images/graemlins/tongue.gif

[johnny005]

my 5ths are 20% over 550 sng's, only 4ths are 12.5%

[Scuba Chuck]

I am beginning to reconsider/retool my game, and to think about the following positive characteristics of a winning player.

1) weak/tight pre-4 handed (unless short-stacked of course)
2) survival equity
3) Avoiding the race

As I think of these three issues, I realize that they are all inter-related or one in the same. Freudian, those numbers are nice. That means nearly 60% of the time, you have an opportunity to make it ITM. Impressive. This will be part of my new challenge. IMO, that can only be achieved by successful weak/tight play, which I think is very proper in tournament games until 4 handed.

For example. I have a few ‘rules’ that I follow. Specifically, I call when I have JJ+, AK. Yesterday, while in the chip lead (t4000), I was dealt AKs when it was 6 handed (blinds 100/200). 2nd stack (t2500), who is on my right pushes allin. I’m multi-tabling. My rule is to call. Reflecting back, survival equity should have pushed me to fold here. There is no need to tangle with this guy at this stage of the game. I don’t have a made hand. As I reflect back to that hand, the correct thought process should have started with this question. “what are the reasons to fold here.” Instead, what went through my mind was, if I win this hand, I’ll dominate the table, and if I lose, I’ll still have t1500 (which put me into 2nd stack position, in a tight race, and eventually out in 4th place).

It's interesting to note the $EV transfer on that call.
t4000 = 32.2% of the equity prize pool
t1500 = 18.2% of the equity prize pool
t6800 = 42.4% of the equity prize pool

I gain 10% of the equity prize pool by winning that hand, but lose 14% if I don’t. Besides the two in the showdown here, everyone else gains by this entanglement. And they gain a lot.

Just some Scuba thoughts I wanted to share.

[EdgePort]

I have been taken this into account a lot more lately, and passing up on a few cases where I would have normally called when fighting with the 2nd stack. An AK example exactly like yours was what made me start thinknig about this. In my case, it was the large stack calling my 2nd stack push and losing that made me realize, it wasnt' really worth the call for him.

[Phil Van Sexton]

[ QUOTE ]

It's interesting to note the $EV transfer on that call.
t4000 = 32.2% of the equity prize pool
t1500 = 18.2% of the equity prize pool
t6800 = 42.4% of the equity prize pool

I gain 10% of the equity prize pool by winning that hand, but lose 14% if I don’t. Besides the two in the showdown here, everyone else gains by this entanglement. And they gain a lot.

Just some Scuba thoughts I wanted to share.


[/ QUOTE ]



Basically, you are saying that risking 2500 chips helps the other players. We know this, and it is already taken into account by the ICM. Just because the other players benefit doesn't mean this is a fold. If you call, it's the raiser that's donating the $EV to the table. He probably has the worse hand AND he'll be eliminated if he loses. That's where the vast majority of the $EV is coming from, not from you.

You have to check AK vs. his range of hands and see if it's a call. You can't just give half the math and let people draw their own conclusions. You seem to be implying that folding AK here is correct. Maybe it is, but you need to show all the numbers before I believe that.

That's a really big raise at the 100/200 level. It would be pretty silly for him to play AA/KK like this.

[Scuba Chuck]

I'm having second thoughts about the math here. I'm thinking that at some stage, particularly this one, considering survival equity (hope I've been using this term correctly) has more weight than a slightly marginal +$EV situation. I know this goes against my math beliefs as well as yours. But there has to be more to this game than just the math sometimes.

And just for math sakes. AKs is only a 45% win probability against any pair. Against just 99+, it's only a 41% win probability. The real problem here is, psychologically at least, what hands would villain push here. I think it's very narrow, AJ, KQs, 99+. These blinds are so low, why would you risk your entire stack?

Furthermore, he did push KK here.

[EdgePort]

I agree, the math may say push or not. But I am not calling a 2500 push to win that 2500 plus 300 in blinds. Atleast not unless I know I am ahead by a good amount. I'd rather keep my chip lead and be able to use that when the blinds get higher to steal more blinds.

[Scuba Chuck]

[ QUOTE ]
I agree, the math may say push or not

[/ QUOTE ]

This is called weak/tight. It's the way this forum views me. And frankly, pre-bubble, they view me correctly. I don't mind. This is a very bad decision, IMO, if it were 4 handed.

[gumpzilla]

[ QUOTE ]
I'm having second thoughts about the math here. I'm thinking that at some stage, particularly this one, considering survival equity (hope I've been using this term correctly) has more weight than a slightly marginal +$EV situation. I know this goes against my math beliefs as well as yours. But there has to be more to this game than just the math sometimes.

[/ QUOTE ]

This reasoning is confused. Equity is equity. What you're saying is that you think that surviving is more +$EV than the slightly +$EV push move, in which case it is mathematically correct to play the way you did. Evaluating the survival equity is harder, but still amenable to mathematical estimation, such as ICM. Saying that you're not sure about the validity of ICM, for example, is quite different than saying that the math doesn't apply. The math doesn't stop being relevant just because you haven't seen the appropriate calculation on here 684 times before. Math is the underlying structure into which you feed all of the other information.

[Scuba Chuck]

Gumpzilla the math here is flawed in only one particular part. And that part is the hand assumptions. For each individual will put what hand they want into the poker calculator, to get an answer. Depending on which one you like, will be your move. The more I think about this hand, the more I think the tighter the hand range. That would make this CALL -$EV. If you want to make the math +$EV, loosen the hand range.

The math is only as good as it's inputs. Don't confuse this too much. Survival Equity is part of considering how solid your read is. The less confident in your read (for this situation only), then folding here could be **correct**.

[gumpzilla]

I agree completely, but this doesn't mean that the math is wrong, it means your read is wrong. Garbage in, garbage out.

One of my pet peeves is when people say that the mathematical approach tells them the wrong thing to do. This only happens when you're providing it with either inadequate or erroneous information, or perhaps using a model such as ICM in a regime where it shouldn't be used. Determining the limitations of such models can be a very difficult question, and there could be places where using it is wrong, but the underlying probabilistic analysis is not at fault there. Perhaps this is all just a quibble over semantics.

[Scuba Chuck]

[ QUOTE ]
Perhaps this is all just a quibble over semantics.


[/ QUOTE ]

This would be a more interesting discussion if the hand was JJ.

[Phil Van Sexton]

[ QUOTE ]

I'm having second thoughts about the math here. I'm thinking that at some stage, particularly this one, considering survival equity (hope I've been using this term correctly) has more weight than a slightly marginal +$EV situation. I know this goes against my math beliefs as well as yours. But there has to be more to this game than just the math sometimes.

And just for math sakes. AKs is only a 45% win probability against any pair. Against just 99+, it's only a 41% win probability. The real problem here is, psychologically at least, what hands would villain push here. I think it's very narrow, AJ, KQs, 99+. These blinds are so low, why would you risk your entire stack?

Furthermore, he did push KK here.


[/ QUOTE ]

I don't know where you got 45% and 41%.

AKs vs QQ (http://www.twodimes.net/poker/?g=h&b=&d=&h=Ac+Kc%0D%0AQd+Qc) = 45.65%
AKs vs 99 (http://www.twodimes.net/poker/?g=h&b=&d=&h=Ac+Kc%0D%0A9d+9h) = 47.43%
AKs vs 22 (http://www.twodimes.net/poker/?g=h&b=&d=&h=Ac+Kc%0D%0A2d+2h) = 49.77%

I don't have PokerStove handy, but it's much better than 45% or 41%.

Pushing with KK here is either genius or cowardice (I think the latter).

Anyway, you gave us a couple more numbers, but you still haven't told us the ICM vs the range of hands you describe.

I don't think you can say "survival equity has more weight than a slightly marginal +$EV situation" because the value of survival is already taken into account by ICM when you computed the +$EV. ICM understands the value of surviving. This is why we use it instead of chip EV.

In addition, you will have 1500 left at the 100/200 level if you lose. You will survive. This is where your "skill advantage" should come into play. Anyone can fold into 3rd place with 4000 chips. A great player takes advantage of a +$EV opportunity if it exists. If you lose, you can use your skill to cash with 1500 chips.

[EdgePort]

4 handed this is a totally different hand, and should be played different.

To the point of the math. It might be that it is +EV to call, and overtime you will win out doing it. But, I will not be in enough of these exact same situtations in my life to make it all even out. Plus my personal playing style and ability also want me to fold this and play my game to steal more blinds while being the big stack, instead of playing as a small-medium stack should I lose.

[Scuba Chuck]

[ QUOTE ]
I don't have PokerStove handy, but it's much better than 45% or 41%

[/ QUOTE ]

I did use pokerstove:

45% is AKs against 22+
41% is AKs against 99+ (actually 41.7%).

I don't think he'd push with pairs lower than 7 for sure. 77 and 88 is the gray area. And, as you're assuming, AA/KK would/might only be mini-raised. If we assume AKs against 99-QQ, then we're still a dog. If you add AQ+, and KQs, then it's a 56% favorite. But I think it's a serious mistake not to include AA and KK in the calculation. Bringing those two hands back in make it a 52% favorite.

So EV fold = 32.2%
EV call = 30.7%

Change HR to include AJ
EV call = 31.7%

So you're right. Thanks for proving your point. I see that ICM does incorporate "survival equity."

[The Yugoslavian]

Just wait until you run into the other side of variance...

Yugoslav

[Phil Van Sexton]

[ QUOTE ]
I did use pokerstove:

45% is AKs against 22+
41% is AKs against 99+ (actually 41.7%).

[/ QUOTE ]


Duh. I filtered out AA/KK. Pretty stupid after you told me he had KK.

I'd expect him to min-raise, 3xBB raise, or limp with AA/KK before pushing. Clearly something I have to think about. I still don't think he'd do this with AA.

So your ICM numbers are for 99+, AQ+, KQs? Very interesting.

[viennagreen]

Eight tabling... finish stats (with a little rounding off)--
1-3 38%
1-4 51%
1-5 63%
1-6 76%
7 9.6%
8 7.5%
9 3.5%
10 3.4%

[viennagreen]

discounting the possibility of AA/KK pushing is definitely a mistake.

i will push with AA/KK in his spot under a couple of circumstances, which occur quite frequently--

1) if i've been pushing a lot, bullying the table
2) if i recently (last 5-10 hands) was caught stealing and doubled up with a poor hand

in either of these situations, i frequently get paid off with my premium hands--- and when i do, it has the benefit of lending my future pushes more respect.

[halis123456789]

you'd be surprised how effective playing AA KK like that can be (depending on your table image)

[Phil Van Sexton]

[ QUOTE ]
So you're right. Thanks for proving your point. I see that ICM does incorporate "survival equity."

[/ QUOTE ]

Since you are trying to translate the ICM into english, I think "survival equity" is the wrong term for this case.

You aren't going to be eliminated here so "survival" doesn't explain why this +EV play is -$EV according to your hand range.

I think a better explanation is: "not all chips are worth the same". The difference between having a 300 chip stack and an 800 chip stack is huge. The difference between 5100 chips and 5600 chips is not huge. In both cases, we are talking about 500 chips. In a side game, there would be no difference.

The ICM is telling us that turning a big stack into a monster stack doesn't buy you that much. Each chip over 4000 is worth less than each chip below 4000.

In this AK example, you are bumping into the ICM ceiling. Due to the prize structure, the ICM will never be greater than 0.50. As you get over 0.40, each additional chip isn't worth much. It's not really a matter of surviving or not surviving.

If you changed the payouts from 50/30/20 to 70/20/10, this would be a call, I think.

[halis123456789]

ScubaChuck, I am viewed as weak tight for a good portion of my tournament life as well. And yes it is a horrible play to call with the AKo here.

He doesn't have anything but Ace high, if the other guy is decent he very well might have AA KK in this spot, even if he has QQ why coin flip (with QQ having the edge) for over 62% of your chips?

Talk math all you want, that's just stupid tournament poker.

[Phil Van Sexton]

[ QUOTE ]
discounting the possibility of AA/KK pushing is definitely a mistake.

i will push with AA/KK in his spot under a couple of circumstances, which occur quite frequently--

1) if i've been pushing a lot, bullying the table
2) if i recently (last 5-10 hands) was caught stealing and doubled up with a poor hand

in either of these situations, i frequently get paid off with my premium hands--- and when i do, it has the benefit of lending my future pushes more respect.

[/ QUOTE ]

Good point. I was thinking (or overthinking) of an opponent with no read.

If the villian were pushing a lot or pushing with weak hands, you should definately include AA/KK, but you should also include a lot of other hands too (Ax?, any pair?).

Therefore, it's still really close for a call. Either you eliminate these hands or you include a bunch of other hands too. Either way, a call may be correct. It's close in my mind.

[halis123456789]

this guy went flat out all in 2500 with 100/200 blinds, he has more than enough chips to last a long long time, he goes all in and you put AQ in the range of hands? in my mind, that's the ONLY hand you can beat here. even 22 has you covered, so why gamble? just pretend you never saw the AK and wait for a better opportunity.