Normal 0

**HILTON NFL CONTEST -
A BAD BET**

© Robert Crowne & Assoc., Sept. 10, 2010

** Note:** The Las Vegas
Hilton Supercontest is a trademark of the Las Vegas Hilton.

The Hilton Supercontest has been peopled, in the main, by

- existing sports services trying to make a reputation for themselves,
- entrants seeking to launch a sports service career based on the publicity they hope to receive, and
- those who pay the $1500 entry fee in the hope of making a substantial profit by finishing in the Top 20.

I've been asked many times why I don't enter the Hilton's contest. The reason is that, in my opinion, the contest is a bad bet, and a poor allocation of business capital.

# What is the Supercontest?

Although the Hilton calls it a contest, the NFL Supercontest is actually a betting pool no different from the pools that might be run by your local sports bar. The "entry fee" is actually a bet by each entrant that he will win more games than every other entrant. The actual number of games won doesn't matter. In theory, the winner could have a losing record overall if everyone else has a worse record. Likewise, someone with a 65% record could lose if 20 of the other entrants finished with higher winning percentages.

The required bet has been set at a substantial $1500. For those investing in the pool to make a profit, the bet is substantial enough to require an examination as to whether the Supercontest is a good bet. For the present and future sports services, that amount is substantial enough to warrant an examination as to whether the Supercontest is a good business investment.

**Betting Aspects of the Contest**

A **good bet **is defined as one at which a skilled
player can gain a substantial enough edge to warrant the risk, or one which
provides a sufficiently substantial edge in the odds. In addition, no matter what the edge, to be a good bet it must be
capable of being won with sufficient frequency to make the risk a good business
investment. For example, if the payout
on a lottery is $20 million to $1, and the odds against winning are 19.5
million -1, the odds provide an edge.
Investing $1500 in tickets would lower the odds against you to 13000-1,
but the probability is still substantially against ever winning in your
lifetime. In such case, despite the
advantage in the odds, risking $1500 in the lottery may not make good business
sense.

The Hilton Contest fails to qualify as a good bet under any of the stated criteria.

**(a) Contest Results Cannot be Handicapped in Advance**

Luck is the enemy of the skilled handicapper. If you have an edge and remove all luck, there is a certainty that you will win the amount of your edge over time. It is only luck in the short term that can make your return uncertain. The skilled handicapper controls whether he wins or loses. The more the skilled handicapper is subjected to luck, the less control he can exert.

It can be said that because the skilled handicapper has an edge, he is like the casino or the bookmaker. He is not gambling; he is in business. He wants certain returns on his business investment. Luck makes the returns uncertain.

For the skilled handicapper, being subjected to luck means being subjected to unknowns that cannot be predicted in advance.

The Hilton Contest adds three elements of luck to the handicapper's equation; and those elements of luck are not merely incidental, but are the main factors in determining whether the contestant wins or loses.

First, in normal sports betting, the proposition is that the bettor will win the individual bet, and overall that he will win more individual bets than he loses. The result of these propositions is capable of being controlled over time by applying one's skill to handicapping the teams.

In the Hilton Contest, the main betting proposition is that you will beat all the other contestants, or at least all but 19 of them. As stated earlier, how many games you actually win or lose doesn't matter, so long as you have more winners than the other contestants. Last year, the contest winner had a 64.8% won-lost record. In other years, it has required over 67% to win.

In the 2009 season, the second place finisher had the same
number of winners as the 1^{st} place finisher, but had the misfortune
of having one game lose rather than push.
That half game difference cost the 2nd place finisher $118,000 in prize money.

Any of the other contestants can beat you with skill, or by
luck, or with some combination of both.
Steve Fezzik is the first person to ever win the contest twice in a
row. To his credit, he attributes his
wins to a great deal of luck. He
has emphasized that the percentage he hit was not a realistic expectation. The effect of luck can be seen in the fact
that Fezzik ranked all the way down in 61^{st} place halfway through
the season, and wound up winning the contest.

The problem for each entrant is that the skill and luck of the other contestants cannot be handicapped. The main betting proposition is a complete unknown. The Hilton Contest situation is similar to betting a horse race without knowing what horses, or even how many horses, are running. In a horse race in which the horses are unknown, every horse must be considered to have an equal chance of winning. Likewise, before the Hilton Contest starts, every contestant must be considered to have the same chance of finishing in any given spot. The final placing in the Contest depends completely on the unknown factors of skill and luck of the other contestants.

Secondly, the contest requires that exactly five sides be picked each week in the NFL only. Since there are 13-15 games every week, the probability is 92.3% that any given handicapper will like more or fewer than five sides in any given week, and only 7.7% that he will like exactly five sides. If the handicapper likes fewer than five teams, he must force the extra selections and will be subject to luck on those selections. If he likes more than five teams, he must eliminate some games and becomes subject to luck of the draw on the five he has chosen.

Thirdly, the exact amount of the prize, and the odds against each contestant are also unknown and therefore subject to luck. The amount of money to be distributed is based on the total number of entries, which cannot be known until after the entry fee has been paid and the Contest is closed. In other words, the contestants are betting blind without knowing the odds against them or the payout odds. In 2008 the top prize was $13,000 higher than in 2009. That's an almost 9-1 difference in odds.

Further, there is a tiebreaker only for the top spot. In 2009, the 3^{rd} place finisher
was only one winner in 85 games behind the 2^{nd} place finisher. If the 3^{rd} place contestant had
one extra winner, there would have been two 2^{nd} place finishers, and
the 2^{nd} place prize would have been divided between the two of
them. If there are more than two people
tied for second, the second place finishers could receive less than the 3^{rd}
place finisher. In that instance, the
final 3^{rd} place finisher would have been moved up to 3^{rd}
from 5^{th,} and receive 100% *more *prize money, solely as a result of
the 3-way tie above him. Whenever a tie
occurs, the results for every bettor at or below the level of the tie are
substantially changed by the luck of the tie.
Ties in the top 20 prize spots are not uncommon.

**(b) The Odds Do Not Provide any Edge **

Since it is impossible to handicap the final position in the contest, the only edge must come from the odds. Many bettors believe that the average $200,000 payout on a $1500 bet, and the fact that all the entry fees are paid out in prize money make the contest a good bet. Nothing could be further from the truth.

As stated above, the most important factor in the contest, the skill and luck of the other entrants, is a complete unknown. The odds must, therefore, be calculated as if every entrant in the contest has an equal chance of winning before the Contest starts.

There were 328 entries in the 2009 contest. Based on that, the odds against any entrant
winning were 327-1. The $1500 entry
fees totaled $492,000.
Normal
0
For the odds of winning 1^{st} place to
have been breakeven, the top prize had to be $492,000. Instead, the winner received only
$196,800 The effect to
the winner is the same as a 60% rake.
The difference between the Hilton Contest and any lottery is that,
instead of pocketing the 60% commission, the Hilton distributes it to
contestants finishing from 2^{nd} to 20^{th}. That gives each of the contestants a better
chance of winning something, but the payout is insufficient in any spot.

Last year's first prize of $196,800, including the contestant's own $1500, represents a payout of only 130-1 on a 327-1 proposition. Below is a chart showing the true odds against finishing in spots 2 through 20, the maximum payout in 2009 for a finish in each spot, and the payout odds represented by the maximum payout. The payout odds in the chart assume there are no ties. :

Finish |
Odds Against |
Payout if no ties |
Payout Odds |

2 |
326-1 |
$78,720 |
52.5-1 |

3 |
325-1 |
$39,360 |
25-1 |

4 |
108-1 |
$19,680 |
12-1 |

7 |
80-1 |
$17,220 |
11.5-1 |

11 |
32-1 |
$ 4,920 |
3-1 |

As you can see, the odds against finishing in any given prize category are substantially higher than the payout odds.

The full payout of fees is reflected in the fact that the odds against winning any full prize at all are 15.4-1, and the average payout odds for all spots is also 15.4-1. On that basis, the Hilton Contest is a no-edge, breakeven luck bet because the house does not cut anything from the pot, but instead distributes the entry fees among 20 or more contestants. The major benefit, however, arises from finishing in one of the top three spots, and the payout does not compensate for the odds against accomplishing that..

The problem arises from the fact that no entry can win more than once, and there is a substantial rake taken by the other winners. Since, before the contest starts, the chance of being in any given position in the top 20 is equal, creating a truly breakeven proposition would require that all the prizes be equal. The varying payouts cause the breakeven to exist only after the same number of trials as there are contestants, Normal 0 or in other words, when the number of contestests entered equals the odds against winning in any given position

In the long term, if a bettor continues to enter the Hilton Contest every year, the bettor would eventually receive every prize and the prize money would exactly equal the entry fees. The Contest is only run once per year, however, and there are an average of 350 entrants every year. Theoretically, winning every prize would take playing in 350 contests over 350 years. The bettor can never reach the long term in the Hilton Contest, and the odds in the short term are insufficient because a rake is taken by other players. Stated differently, the bettor will not live long enough to be the recipient of his full share of the rake. If the payout were equally divided among 20 winners, then the payout would always equal the odds of winning a prize, and the bet would be a breakeven proposition in the short term as well as in the long term. That, however, is not the payout structure.

Another negative factor arises from the fact that there can be an unlimited number of ties in
any spot except first place. In theory,
after the first place finisher is determined, all remaining contestants could
finish 2^{nd} through 20^{th}.
Last year they would have represented 327 contestants, or 99.97% of all the entrants finishing 2nd through 20th.

If that had happened, after the 40% was paid to the winner, there would have been only 60% of the pool remaining to be paid to 99.97% of the contestants. That means each entrant would be guaranteed to receive less than they bet. Winning and receiving less than you bet could never happen in a pari-mutuel pool or at the bookmaking window.

The problem lies in the fact that an unknown number of winners must be paid from a fixed pool, and the percentage of the pool taken by the first place winner is fixed. The result is that even after winning every prize over 350 years, a bettor still may not receive a return of all his entry fees due to the occurrence of ties.

**(c) The Expected Frequency of Winning is Insufficient**

Even though all contestants could, theoretically, qualify
for a prize, don't count on getting any prize at all, even less than you
bet. Last year you would have had a
94% (308 in 328) chance of losing your whole $1500. The odds against getting the top prize are 327-1. Since the contest is run only once per year,
the player can expect to finish first once every 327 years. The odds against a 2^{nd} place finish and a 3^{rd}
place finish are the same as for a 1st place finish.

Normal 0

That two people have managed to win the Hilton Contest more than once means no more than that a woman hit the Megabucks slot jackpot twice in two days, and a man won the top prize in the NYS lottery twice in two years. Winning the Hilton Contest twice is a 107,000 to 1 shot. If 107,000 to 1 shots never came in, no one would every win a State Lottery. Those multiple-win occurrences are a part of random selection, and their existence does not make the slot machines, the NYS Lottery, or the Hilton Contest good bets, nor prove that they are games governed mainly by skill rather than by luck. .

In the hope that their one win in 327 years would come in 2009, contestants took the 94% chance of losing everything despite having no advantage in the odds, and no advantage that could be calculated in advance based on skill. That's not a risk I believe to be smart.

The question remains, whether the risk at breakeven is worthwhile for a sports service as a business investment for the publicity.

# Business Aspects of the Contest

There at two possible benefits for current or future sports services. Those are, the creation of an independent record, and the possibility of free publicity. if the service wins. The latter is probably the more important consideration.

There are many less expensive, more complete, and more accurate ways of documenting selections than paying a $1500 contest entry fee. The limitation of the contest to NFL sides, the requirement that there must be exactly five selections every week, and the requirement that those selections be made at least two days before the games are played, makes the contest an inaccurate measure of the handicapper's skill.

Additionally, in the final weeks, the selections for those on the bubble
must be based on contest considerations rather than handicapping. In 2009, Fezzik was still attempting to
catch up in the final week. The only
way to be sure of overtaking the leader was to pick opposite him and win. The only way for the leader to be sure of
remaining in front was to try to pick at least three sides that were the same
as Fezzik's picks. During an interview with
the Las Vegas Sun,. last year's winner, Fezzik, said, "I data mined my opponent
and noticed that he played a lot of four-to-seven point underdogs." Based on that Fezzik believed the leader at that point,
Big E, would pick Seattle, and Fezzik chose Seattle's opponent, Tennessee. Big E did take Seattle, and Tennessee got
Fezzik 1^{st} place by ½ game.
End-game picks in the final weeks have less to do with handicapping
skill than with tournament skill.

From a publicity standpoint, there may be some benefit to a
finish in one of the top three spots, but the real publicity value comes from
being #1. Those who desire to enter the
sports service business on the strength of their Hilton Contest performance
will quickly discover that the public doesn't care about a 10^{th}
place finish. Why pay the guy who
finished 5^{th} when you can pay the guy who finished first or
second? If the 1^{st} place
winner is not already in the sports service business, you can bet he'll be getting offers from
the big boiler room operations.

Further, only the winner gets free publicity from the news media and the Hilton. Do you know the name of the #2 and #3 finishers from last year or the year before? A handicapper who does not finish first will need to spend just as much money advertising his lower-place finish as he would need to spend if he had never paid the $1500 to enter the Supercontest.

If I spend $1500 on well-chosen conventional advertising, I am certain of a profitable return. If I spend $1500 on a 327-1 chance that I will get free publicity from the Hilton Contest, I have a 99.97% chance of getting nothing for my money. Entering the Hilton Contest instead of buying conventional advertising is not a wise business decision.

# A Creative Solution

I have arrived at an idea so simple that I can't believe it took 20 years to dawn on me. I can enter the Hilton Contest without violating my principles by making a bad bet, get all the same business benefits I am likely to get from the Hilton Contest, and make money at the same time. I don't need to donate prize money to 19 or more other Hilton contestants, and I don't need to pay $400 plus 1% of any prize money to a proxy in order to have my selections entered for me in Las Vegas every week. I also don't need to spend time and money traveling to the Mojave desert in the 130 degree heat of August to enter the Contest in person, as required. Instead, I can post my five Hilton selections every week right here at Pregame. Whether I put the selections up free in the forum, or in a pay package, will depend on several factors including how I've been doing, and how I feel about the selections that week.

Periodically putting the selections in a package will help to avoid the problems that arise from publicly posting selections every week that members of the Crowne Club have paid to get, and it will help me to earn some money from my effort. The benefit of the pay packages over my free posts and over the selections of others in the contest is that I will rate them so that you know which plays are real, which may be weaker or simply forced, and toward the end of the contest, which selections may merely be strategic.

A substantial number of weeks will be free. I will keep track of the free selections and the posted selections in the forum, and of course you can track the free selections yourself and the pay selections by going to my page http://pregame.com/pregamepros/pro-bettor/bettor.aspx?id=4441#daily_msg

If I happen to finish with a record that would have placed me in the top three places in the Hilton Contest, you can all attend, free of charge, as a I cut off my own head with a ceremonial samurai sword in the parking lot outside the door to the Hilton Sportsbook. Refreshments will be served.

If I finish anywhere in the top 20, I would be forever grateful if all of you thank me by going forth on the Internet and publicizing my accomplishment in as many places as possible..

If this is Friday or Saturday, the picks are posted in a thread in the NFL section of the forum under the heading "Rob Crowne Hilton Picks" or in a package on my pick page. Picks will be posted by the Saturday, September 12, 2010 Hiltoin Contest entry deadline, and by the deadline each week thereafter.

__ Disclaimer: __ My
selections to Crowne Club members and in late packages are often based on
professional syndicate action, or on late game information including weather
and injuries. My selections for the
contest may be made for strategic tournament reasons rather than handicapping
reasons. Thus, my posted Hilton
selections may be different from or even opposite my service selections.