This looks like an investment with a 17% advantage. Read the details at cardplayer.com. As usual, the article doesn't tell the reader the important information to determine the value of shares on Amir. Here is how many chips each player has and the share of the total.

Player | Chips | Share |
---|---|---|

JC Tran | 38,000,000 | 19.93% |

Amir Lehavot | 29,700,000 | 15.58% |

Marc McLaughlin | 26,525,000 | 13.91% |

Jay Farber | 25,975,000 | 13.62% |

Ryan Riess | 25,875,000 | 13.57% |

Sylvain Loosli | 19,600,000 | 10.28% |

Michiel Brummelhuis | 11,275,000 | 5.91% |

Mark Newhouse | 7,350,000 | 3.85% |

David Benefield | 6,375,000 | 3.34% |

Here is how much each place wins.

Place | Win |
---|---|

1st | $8,359,531 |

2nd | $5,173,170 |

3rd | $3,727,023 |

4th | $2,791,983 |

5th | $2,106,526 |

6th | $1,600,792 |

7th | $1,225,224 |

8th | $944,593 |

9th | $733,224 |

Total prize money to top 9 = $26,662,066. Amir's expected win above the guaranteed 9th place win is 15.58%*$26,662,066 - $733,224 = $3,419,724. Each 1% of that is worth $34,197. The article says each 1% share costs $29,248. That would be an expected return of 34197/29248 = 117%, or a 17% advantage.

Assuming he can be trusted to pay, and not throw the game, this is a great value.

Quote:BozLooks like Amir doesn't have a lot of faith in his ability to finish where he is currently at. Or am I looking at this too simply?

He is simply hedging. If he sold all of his shares he guarantees $2.9 million. Bad luck/bad play could limit him to 700k. If he wins the main event, even though he is out many millions initially, he gets entries into other big +EV events, endorsements, etc.

It's a question we ask here periodically, when does a number become big enough that chasing more +EV is not worth it? I think we would all take 100% chance at $2.9million over a 10% chance at $32million as an example.

Quote:DRichThis seems like a win/win for Amir. Sell 300% of yourself, bust out first, make $10 million.

Springtime for Hitler?

Quote:kenarmanIt seems like a good deal to me. Amir is a poker pro and has won a $10K buy-in tournament before so knows what he is doing.

How are his past results remotely relevant if not taken into the context of his opponents results?

He could be the 9th best poker player on the planet. If the other 8 are at the table ...

Quote:onenickelmiraclehe would have no motivation and I don't think he would want to win if he sold all his shares. To me it sounds like a foolish investment unless there is some proof of him having character.

I think there is plenty of motivation for him to win even if he sold all the shares. The bracelet and all the benefits that go with it are worth the few extra hours of play.

Quote:onenickelmiracleThe tempting thing for him in a sold-out scenario would be to go all-in fast, so he is guaranteed a profit on the sales. After a double-up, he would want to stay in to place higher. Winning it all would wind up being more luck, since he would have no motivation and I don't think he would want to win if he sold all his shares. To me it sounds like a foolish investment unless there is some proof of him having character. It would really wind up depending on how much of his own stake he is still invested in to keep him motivated to try to win as much as possible.

I find it difficult to imagine that he should want to lose.

How about this: Maybe he REALLY wants to win, so much so, in fact, that he is concerned that thinking too much about the money will distract him. If he sells all of his shares, and has a fixed profit, then he can play his game with a nice clear mind!

Quote:thecesspitWe assume he wants to sell all his shares. Is that whats he's said? Selling 20% might make a lot of sense for risk mitigation.

No reason to assume, the article specifically states that he is selling 30%, keeping 50% himself, with the other 20% going to another investor and his parents. So it is risk mitigation.

Quote:WizardThis looks like an investment with a 17% advantage. Read the details at cardplayer.com. As usual, the article doesn't tell the reader the important information to determine the value of shares on Amir. Here is how many chips each player has and the share of the total. ...

Total prize money to top 9 = $26,662,066. Amir's expected win above the guaranteed 9th place win is 15.58%*$26,662,066 - $733,224 = $3,419,724. Each 1% of that is worth $34,197. The article says each 1% share costs $29,248. That would be an expected return of 34197/29248 = 117%, or a 17% advantage.

Assuming he can be trusted to pay, this is a great value.

I found my way here after listening to the podcast and starting a thread in that forum. I'm surprised at this analysis after the article itself mentions ICM. A 17% advantage, assuming equal skill, is wrong. He has 15.58% of the chips, and may have a similar % chance of obtaining all of the chips, but he does not have that chance of obtaining all of the prize $. If you think he's a better player than the opponents, then you have an edge, but you shouldn't think he mispriced the chips apart from his skill, which is exactly what the article says.

John

Quote:socksHe has 15.58% of the chips, and may have a similar % chance of obtaining all of the chips, but he does not have that chance of obtaining all of the prize $.

Why not?

Quote:WizardWhy not?

I think this is just semantics. No one can obtain "all of the prize money". Getting all of the chips just gets you the first place prize money.

Quote:SOOPOOI think this is just semantics. No one can obtain "all of the prize money". Getting all of the chips just gets you the first place prize money.

I'm not saying that he can. I am saying he can expect to win 15.58% of the total prize money available to the top 9. If anyone disagrees, let's see your math.

Quote:WizardI'm not saying that he can. I am saying he can expect to win 15.58% of the total prize money available to the top 9. If anyone disagrees, let's see your math.

This is absurd. The burden of proof is on the person making the constructive statement that lacks any logic, not the person saying that the person is without any logic. That being said, poker players have been using icm for many years and it's one of the main secrets to success among people that make money in single table tournaments. There is no constructive logical reason why tournament chip standings should be proportional to eventual cash winnings. Doubling your stack does not double your equity if the tournament is not winner take all, and people have been using that to make money hand over fist for years.

Here is a simple empirical test. Run a regression of eventual main event winnings on final table chipstack over a period of many years and players. Is the coefficient equal to 1? In other words, lets just see how your investment strategy performed over the past two decades of final tables and how much money you would have won or lost.

Quote:randompersonThis is absurd. The burden of proof is on the person making the constructive statement that lacks any logic.

Although I consider your remark very rude and insulting; I concede I may have been in error. I'm working on a more robust model for the calculation. A more effective and polite way to refute my theory would have been a case where each place paid the same amount.

Quote:WizardAlthough I consider your remark very rude and insulting; I concede I may have been in error. I'm working on a more robust model for the calculation. A more effective and polite way to refute my theory would have been a case where each place paid the same amount.

I tried that earlier in the thread, it didn't seem to work.

A more obvious example may be someone that has 80% of the chips when first place pays 50% of the prize pool. Clearly that person's equity is not only less than 80%, but also less than 50% of the prize pool. Now just think about whether that effect displays some sort of continuity and there you go.

I'm not saying that he can. I am saying he can expect to win 15.58% of the total prize money available to the top 9. If anyone disagrees, let's see your math.

The problem is that the final answer lies not in math, but in that person's ability. Perhaps a bigger factor is how much he is willing to risk to win first prize money.

Think of it this way.

AMIR leaves the table and WSOP allows me to take his place. What is my equity ? If you say it's the same as AMIR's , you are so WRONG ! ! ! ! ! ! ! !

And how big a roll would ICM or BOB or whatever have in the other player's decision making. Less than ZERO. Even if we say ICM

is 100% against AMIR, then how about all the players between our skill sets ?

ICM, 30 Big Blinds, Always open for 3 times the BB, all of this only works if you believe in it. If you win HOORAY. If you lose, well. I made the right move. How in the hell did anybody ever win before ICM. Boogles the mind.

Quote:WizardAlthough I consider your remark very rude and insulting; I concede I may have been in error. I'm working on a more robust model for the calculation. A more effective and polite way to refute my theory would have been a case where each place paid the same amount.

It is frustrating when you offer advice w/o even acknowledging the tools of the trade which are mentioned in the article you took the problem from. ICM has been accepted to so long in tournament poker circles that it is generally the answer, not the question.

If you want more depth, you should go to The Mathematics of Poker by Bill Chen and Jerrod Ankeman. According to wikipedia, Chen has a PhD from Berkley and heads the StatArb department at Susquehanna International Group. It reads almost like a testbook and Ferguson is quoted on the back as saying that if he taught a poker math course, this would "be the only book on the syllabus."

Ch27 - Chips Aren't Cash, starts with "A Survey of Equity Formula's", including "The Proportional Chip Count Formula" which, at a glance, looks similar to what you were doing, as well as 3 others: The Landrum-Burns Formula, The Malmuth-Harville Formula, and The Malmuth-Weitzman Formula. Some googling shows that ICM is The Malmuth-Harville Formula. Chen/Ankeman say that "The Landrum-Burns and Proportional Chip Count methods have clear flaws but are easy to calculate at the table. ... the Malmuth-Harville and Malmuth-Weitzman formulas are more accurate, but harder to implement."

John

I contend that my original methodology would be right in a winner take all tournament. However, the WSOP structure of diminishing prizes gives the smaller stacks at the table a greater than chip-proportional share of the prize pool.

My new methodology assumes that chips keep randomly moving back and forth between random players, unless a player gets knocked out. All players are assumed to have equal skill. The math calculation is a simple brute force looping program. I've done this before for a glorified Derby game I did for a client years ago.

So, here is my new poker calculator. Just put in each player's stack size and the prize amounts, click calculate, and voila, you'll see each player's probability of finishing in each place, and the expected prize money on the bottom.

After putting in the information from my initial post on the topic you'll see that Amir has an expected win of $3,658,046. Then subtract out the minimum prize of $1,647,843 for 9th place and you get $2,924,822. Each 1% share has a value of $29,248.22. This is conveniently the price quoted in the cardplayer.com article.

My apologies for the original error.

Quote:WizardOkay, pass the crow and the humble pie. My original answer was wrong. I'll also ask Bob to read a retraction on the air this Thursday.

Wizard I want you to know that the cute/funny things you say always make me laugh.

ALL HAIL THE WIZARD!!!

Quote:OzzyOsbourneWizard I want you to know that the cute/funny things you say always make me laugh.

Thanks! Now if only the women would say that.

Quote:Wizard

So, here is my new poker calculator. Just put in each player's stack size and the prize amounts, click calculate, and voila, you'll see each player's probability of finishing in each place, and the expected prize money on the bottom.

After putting in the information from my initial post on the topic you'll see that Amir has an expected win of $3,658,046. Then subtract out the minimum prize of $1,647,843 for 9th place and you get $2,924,822. Each 1% share has a value of $29,248.22. This is conveniently the price quoted in the cardplayer.com article.

My apologies for the original error.

FYI, some of the places are not written appropriately on the calculator:

For example Player 3 looks like this:

Player 3:

Place 1 = .

Place 2 = .

Place 3 = .

Place 4 = .

Place 5 = .

Place 5 = .

Place 5 = .

Place 5 = .

Place 6 = .

Another issue is: Leaving a stack blank or inputting zero into a stack breaks it. So need to have some type of flag to reduce the number of players in the game when either their chip stacks are blank or zero.

9th place pays: $733,224

$3,658,046 - $733,224 = $2,924,822

I have avoided page 1 of this thread because I am trying to not learn the identities of the entire October 9...lol

Good job doing this in little time at all.

I tested a sample versus another ICM calculator, and all equities matched, so looks pretty good other than that zero chip stack bug and the copy-paste issues with the Place results for each individual player.

Quote:tringlomaneFYI, some of the places are not written appropriately on the calculator:

Thanks, I fixed that.

Quote:Another issue is: Leaving a stack blank or inputting zero into a stack breaks it.

Yeah, I know. Until I find a fix for that please put in 0.000001 for any zero stack, to trick the calculator. There is a division by 0 error for 0 stacks, which I'm having a hard time trying to fix.

Quote:WizardOkay, pass the crow and the humble pie. My original answer was wrong. I'll also ask Bob to read a retraction on the air this Thursday.

So, here is my new poker calculator.

Thank you.

That's a fast turn around on getting the calculator up.

Quote:Another issue is: Leaving a stack blank or inputting zero into a stack breaks it. So need to have some type of flag to reduce the number of players in the game when either their chip stacks are blank or zero.

Okay, it is working better now. If there are fewer than nine players you have to start with player 1 and keep going, without skipping players. For example, if there are six players in the tournament enter amounts for players 1 to 6. The program will get confused if you skip over positions.

Quote:WizardOkay, it is working better now. If there are fewer than nine players you have to start with player 1 and keep going, without skipping players. For example, if there are six players in the tournament enter amounts for players 1 to 6. The program will get confused if you skip over positions.

Expected win is still busted unless zeros are placed in all of non-paying payouts (blanks give NaN). And also non-participants get non-zero probabilities for the places not used, which is problematic if someone overflows the payout table with non-zero payouts. It is better though!

Quote:tringlomaneExpected win is still busted unless zeros are placed in all of non-paying payouts (blanks give NaN). And also non-participants get non-zero probabilities for the places not used, which is problematic if someone overflows the payout table with non-zero payouts. It is better though!

Thanks. I just monkey-proofed it some more. Please try again.

Quote:tringlomaneFYI, the final table is ~live on ESPN2 tonight starting at 5PM PST (final 3 play tomorrow).

Thanks for the reminder. I have about $800 on JC Tran NOT to win.

Quote:WizardThanks for the reminder. I have about $800 on JC Tran NOT to win.

What odds you get for that one? JC is an okay guy but never been one of my faves to watch. If a virtual unknown gets you money, let's go no namers!

Quote:WizardThanks for the reminder. I have about $800 on JC Tran NOT to win.

What is the return? Seems like the right bet.

Quote:BozWhat is the return? Seems like the right bet.

Generally speaking I agree. I would have to get pretty bad odds to not take it.

Quote:BozWhat is the return? Seems like the right bet.

I have some at -346 and some at -375.

Quote:WizardI have about $800 on JC Tran NOT to win.

A very good bet. But Al and me are pulling for Tran. We want to see the old guard knock off the whippersnappers. However, at this time, Tran has fallen back to 4th chip position with 6 players remaining. Go Tran!