Discussion: "Do I really need to push All-In"

[DavSimon]

Here is a link Soupie posted, I figured it would make for an interesting discussion. It got Soupie thinking….and it certainly got me thinking about the ideas put forth. What follows is a little discussion between Fnord and myself, I was wondering if anyone could add anything…experiences – thoughts – even data, like if you won a R&A where were you chip-wise at the end of the rebuy and how much you invested. I have been close several times and regularly squeak into the money or bubble out just before, so my style of play seems to be working fairly well, I just have not connected when it mattered, so this really got me wondering if I should simply save some money in the rebuy period and play it like a freezeout (does a early big stack actually help you, or does it simply make you more comfortable) Yes I do realize that feeling comfortable or safe has a positive effect on your game, so there is the “state of mind” aspect to a big stack, but over-confidence can lead to playing too many hands…simply because you can afford it.

http://www.rakefree.com/faq1.htm#faq

Fnord's response to my comments:

Very interesting, thanks for the link. I wonder if he means all-in as in every single chip or he would include situations where you would be required to invest ~3/4 of your stack - to your opponents AI bet. If you win your are in good shape, if you lose you are pretty crippled when compared to the blinds and antes. Essentially, you would still be in the tourney, but you would be operating at a fairly large disadvantage....and then you would be forced to get all-in once or twice to get back to a competitive stack (I guess I answered my own question)
Very thought provoking, definitely worth keeping in mind while you play to see if it would put you at a disadvantage -or- if you opponents start to pick up on the fact that you seem to be avoiding committing all your chips and start to take advantage of you. Kinda flies in the face of how I have been playing during the rebuy period recently....essentially loose and wild, trying to get as many people to push with me as I can. It has been working, the last two days I made it to the first break with 11K and 19K respectively. I am usually shooting for somewhere between 8K-12K....hmmm...now you got my mind racing, thinking about the possibilities/implications

Consider:

It costs $31? for a ~5k stack if you just camped hands the first hour. How much are you paying in re-buys for your 10-20k stack and how much is it increasing your chances to win? TPFAP talks about the nth chip being worth less than the nth+1 chip...

More food for thought, no one who has had the chip lead after day 1 of the WSOP main event has ever won it. -Fnord



Here was my response:

It varies from tourney to tourney, quite often it only costs me the same $31 to get the 10-15K in chips (excellent value) but the most I have ever spent is 5 rebuys and 1 add-on...so in a $10 R&A, like yesterday I spent $61 to get 19K...still and excellent value, since I only invested enough to "purchase" 5x1500 & 1x2000 The money I invested should have only given me 9500 chips....essentially a 200% return on investment (does not always work out that well)

I have never been the chip leader after the first hour, I am usually in the top 10-15% though. But the point is well taken - I have never come in first place, but I do finish in the top 5-15% very consistently. The true question is (I would like to know) how much does that above average stack increase my ability to weather the trials of hour 2 and 3. I don't recall seeing the chip leader after the first hour winning the tourney...but I have also not seen the person who went into the second hour with 5K win either. There has to be an optimal or close to optimal stack size that maximizes value and increases the opportunity for you to be in a position to win. Does anyone know what that is? I suppose it would take a lot of data and number crunching - analyzing what the initial $ invested and chip stack size after the rebuy period of tourney winners is

In conclusion: I suppose live tourneys are significantly different that online tourneys in that the blinds increase so much faster….you are basically required to get all you chips in the middle several times to simply survive, let alone make it to the top 10% of the field late in a MTT. However I think we/you/I tend to act too quickly sometimes and push all-in when we really didn’t need to do it….putting ourselves in a position to take a big hit or even worse, end our tourney right there. Anyway, think about it while you are playing today…this weekend and ask yourself do I really need to push here or will a 3-5xBB raise have the same effect.

Comments and discussion is absolutely welcome (comments regarding your opinion of Dutch Boyd are not welcome) Lets try to keep this an on topic discussion….thanks for reading - Dave

[Sed]

If you guys can throw out enough numbers like how many rebuys, what your stack was at the first break, madness or tight play, and what you placed, I am sure that Zenbitz and I can come up with a statistical analysis of how much more/less likely you are to win being a maniac the first hour.... I would guess we would need over 100 samples for this to work out.

I think that a slightly different question that should be asked is if it is any less likely to win with the minimum after the rebuy period is over. Essentially if you bust out the last hand, rebuy 2X and add on at the end.

- sed

[zenbitz]

I was thinking that someone must have done some basic work on the statistics of tournament poker. Well, I found this... kinda funny

http://www.twoplustwo.com/digests/to..._msg.html#9521

Something doesn't make sense about what Dutch Boyd was saying... the reason he won (well, came in 2nd or whatever) was that he never committed all his chips. Yeah, sure, great plan. Unless every time you catch something, someone ELSE raises you for all your chips. But perhaps most of the WPT guys are very conservative as well, and respect bets.

That's really besides the point. The question is - we all know that the 2nd 1000 chips are not worth the same as the first ... or do we? I will have to think more about this. I would be some simulations might be in order.

More thoughts from Greg Raymer and Co, 2+2, ca. 2001.
http://www.twoplustwo.com/digests/to..._msg.html#9486

[whileone]

This sounds a bit like marginal value and marginal utility
http://www.daviddfriedman.com/Academ...Chapter_4.html
The gist is, when a particular resource is plentiful, it's value goes down. The example is, if oranges cost five dollars each, I won't make orange juice, but if they only cost ten cents each, I will make orange juice.

So, the less valuable a resource is the more uses it is put to. If you have a big stack, you can make looser and looser calls, because the value of busting someone out exceeds the value of the chips in front of you.

[zenbitz]

Heh, I know David Friedman from the SCA (medievel "recreationsists")

But there are actually two forces here.

1) The value of addition chips (for you) is lowered as you get more. "The 2nd thousand is not worth as much as the 1st". This is because when you bust out you are done.

2) The value of chips goes DOWN as the tourney goes on. We do this implicitly all the time when we 'scale' our chips in terms of the BB. So, 2000 chips when the blinds are 10/20 are worth roughly 1/10th when the blinds are 100/200.

Now, to the question of rebuys - what is the value of those chips in real dollars? I believe mathmatically, rebuying chips while pushing with marginal holdings is a fool's errand. Unless you are getting better than 1:1 on your all-ins (i.e. coinflip+), then you are better off just entering more tourneys. (This assumes that you, unlike me, have time to play multiple MTTs). Note that this assumes that all players are equal.

HOWEVER, the last statement is patently false. We are not playing math, we are playing poker. We know poker players, particularly at the beginning of a large tourney are NOT equal. What you are doing by going ripptyde on your table in the beginning of the rebuy is getting into your opponents' heads. As long as you are getting some folds (particularly with multiple limpers when blinds are small), AND forcing people to make bad/marginal calls later, then the pay off is there.

The problem, I seriously doubt this is calculatable, even with a very sophisticated poker simulation. It would have to have an AI that would "learn" your behavior and act INAPPROPRIATLY. "Artificial Stupidity".

Well, it would make a cool project.

[TylerK]

Heh, I know David Friedman from the SCA (medievel "recreationsists")

You know you just turned this from an interesting discussion into 50 people replying "Lightning bolt! Lightning bolt!" at you, right?

[zenbitz]

Well, I have been a nerd for 20-odd years, I think I can take it. The difference is that in the SCA we beat the crap out of each other with real sticks.