The biggest potential problem is that handcapping is a feat of skill, so it's actually possible for someone to beat it. My idea of dealing with that is that I make the challenge available only to tout-picking services (those that actually sell their picks). Anyone who's really good at handicapping can make way more money betting the games themselves than selling their picks, so they generally don't sell. I presume those who are actually selling don't have a great chance of actually picking a winning season.

From there, the question becomes, assuming no special skill and random outcomes, what are the odds of picking a winning season? Let's say that the challenge is to pick a majority of winners over the 332 games in an NFL season (i.e., ≥167 winners). If the house edge is 4.54%, then I presume the chances of picking a winner in any particular game is (100% - 4.54%) ÷ 2 = 47.73%. From there, I'm stumped as to how to continue the calculation. I found this formula which I think applies (calculates the chances of winning x times out of n trials), but Excel chokes when trying to compute a factorial of 332.

I wrote a program to simulate it, and found that the odds of picking a winning season are around 18.7%. But now i'm wondering if there's a way to calculate it, given the problem with Excel.

Sooo...your thoughts please on:

(1) Whether I should offer the challenge (and why or why not).

(2) If so, how I should structure it.

(3) How to calculate the odds of having a winning season?

Below is the WoV post (now deleted) that got me thinking about this.

Quote:some_loserOUR NFL SYSTEM WILL TEAR THE COMPIITION APART AND HAVE THE BOOKIES ON THEIR KNEES BEGGING FOR YOU TO GO AWAY...WE HAVE A GUARANTEED SYSTEM THAT'S PROVEN TO WORK...GIVE US A TRY...YOU CAN'T LOSE...[website address redacted]

********2016 NFL SEASON EARLY BIRD SPECIAL********

WE ARE RUNNING AN EARLY BIRD 2016 NFL SEASON PACKAGE FOR ONE WEEK ONLY FOR JUST $599.99. THIS PACKAGE WAS EXCLUSIVE TO OUR CUSTOMERS ONLY BUT, WE’VE DECIDED TO OFFER IT TO THE PUBLIC IN HOPES OF GAINING NEW CLIENTS. THIS PACKAGE INCLUDES ALL PLAYS FOR THE 2016 NFL SEASON, PLAYOFFS, & SUPER BOWL. DON’T MISS OUT ON THIS DEAL BECAUSE BEFORE THE SEASON STARTS OUR NFL SEASON PACKAGE ALONE WILL COST YOU $1,999.99. YOU WILL SAVE A TON OF MONEY BY GRABBING THIS DEAL RIGHT NOW. WE WENT OVER 70% ON THE 2015 NFL SEASON SO BELIEVE ME IT’S WORTH EVERY PENNY! THIS OFFER WILL OFFICIALLY START AT 8:00 P.M. 4-26-16 AND WILL END AT 11:59 P.M. 5-3-16 & WILL NOT BE OFFERED AGAIN! I’M SO CONFINDENT IN MY CAPPER’S ABILITIES THAT I’LL OFFER YOU THIS AS WELL “ IF YOU BUY THIS PACKAGE & YOU HAVE A LOSING SEASON IN THE NFL, WE’LL GIVE YOU THE ENTIRE 2016-2017 MEN’S NCAAB SEASON FOR FREE ”! THIS IS A BLOWOUT DEAL NOBODY ELSE CAN OFFER! PLEASE VISIT US AT TO PURCHASE THIS DEAL.

If you notice the ad you quote says 'we went over 70%.....'

Make your challenge simple..... You must pick at least 100 games and score at least 60%. One problem you will have is determining what the real available line for any game is. And of course, no money line games....

Quote:MichaelBluejaySome yahoo is spamming the forum to advertise his pricey NFL picks. That got me thinking, maybe I should set up a challenge for sports handicappers, similar to the Betting System Challenge. The idea would be to debunk the touts -- not because they'd fail the challenge, but because they'd never submit to the challenge in the first place.

That is definitely an interesting concept, Mr. Bluejay, but as you have already highlighted the main difficulty with your Betting System Challenge is that it is patently not winnable due to the House Edge of the game. For an endeavor such as this one, you would actually have to come up with something that on its face offers better Odds than the Sports Book, (if this is to be an Even Money bet) but obviously you want it to offer worse than a break even chance of success or its not a good bet for you.

Furthermore, even if you did come up with such a structure, there's going to be much more exposure to you than there is with a Betting System Challenge. With the Betting System Challenge, you essentially can't lose.

Quote:From there, the question becomes, assuming no special skill and random outcomes, what are the odds of picking a winning season? Let's say that the challenge is to pick a majority of winners over the 332 games in an NFL season (i.e., ≥167 winners). If the house edge is 4.54%, then I presume the chances of picking a winner in any particular game is (100% - 4.54%) ÷ 2 = 47.73%. From there, I'm stumped as to how to continue the calculation. I found this formula which I think applies (calculates the chances of winning x times out of n trials), but Excel chokes when trying to compute a factorial of 332.

I'm afraid that your presumption seems to be in error, as you can see from the Wizard:

http://wizardofodds.com/games/sports-betting/

When it comes to Totals Bets or Point Spread bets, the probability of merely winning the bet is presumed to be about 50%, where the House Edge comes from is the vig, which is to say that a player lays eleven units to win ten units on the game. With that said, a simple Binomial Distribution will tell you the probability of having 167 (or more) winners from 332 games:

http://vassarstats.net/binomialX.html

That results in a 47.8% chance of picking 167 winners (or more) out of 332 games assuming a 50% probability of winning on any given individual pick. Of course, the result would still be a losing season because the results would be:

(165 * -11) + (167 * 10) = -145

In other words, you would make 3652 in total units wagered and would lose 145 units. Therefore, the return is:

(3652-145)/3652 = 0.9602957284

In order to produce a winning season making 332 Point Spread or Totals Bets, a player would need to have 174 wins against 158 losses for a profit of 2 units. That represents a winning percentage that rounds to 52.41%...and 52.4% winners is close to the cut off as you can see:

(47.6 * -11) + (52.4 * 10) = 0.4

We will use the same Binomial Distribution to figure out the probability of Picking 174 winners of 332 and we get 20.521% to pick 174 winners (or more) in 332 games.

You should also recall that this success rate in excess of 20% does not reflect any skill taken in picking whatsoever, it could just as easily reflect a monkey trained to either point to the right or the left and picking whatever side he points to.

Furthermore, there are very few tout services that would suggest that they are going to pick every game throughout the entire season, so as a result, most such services are actually going to have a greater probability of overall success for a season (even if the picks are random) due to a smaller number of trials.

In my opinion, it would be an interesting challenge to structure, but it would be unfeasible to come up with a uniform set of Rules. The only way that I can see this working is if you had a challenge in which a certain number of Totals or Point Spread Bets must be made throughout the course of a season, that the tout must flat-bet every single game, and obviously you would have to determine how much money would be won after the vig is taken out.

The problem is that the House Edge on Totals and Point Spreads is, in fact, relatively low and you would have to structure a Challenge (with your Even Money bet of 10K) that actually gives the potential challenger a lower House Edge (you being the House) than he is getting on the sum of his bets. I'm not saying that you have to do that in order to be generous, I'm saying you would have to do that or the tout has good reason to refuse your bet.

In the Westgate Superbook contest half of the players had a .500 or better record.

With the idea of keeping it simple, I like SOOPOO's idea of picking 60% of 100 games. It's easy for the public to understand, not onerous for touts, and we don't have to screw around with determining what the lines are. The chances of winning with this structure is less than 3%, if I figured correctly.

This would exclude bona-fide skilled touts who could pick 52.4-59.9%, but skilled and non-sleazy touts aren't the ones I'm trying to debunk. I'm trying to debunk the ones that promise a 70% winning season and that "you can't lose".

However, if 100 games over a season is more than most touts would pick, then maybe 75 games is more reasonable. With 75 games, the chances of picking 60% winners would be around 5%.

Howzat?

By the way, I have to comment on this part of the spammer's come-on: "IF YOU BUY THIS PACKAGE & YOU HAVE A LOSING SEASON IN THE NFL, WE’LL GIVE YOU THE ENTIRE 2016-2017 MEN’S NCAAB SEASON FOR FREE"!

In other words, "After you lose money after paying for our worthless picks, we'll make it up to you by giving you some more worthless picks for free." And note the exclamation mark, like he's really excited about that.

Sadly, I think a lot of people would fall for it.

Quote:MichaelBluejay

The biggest potential problem is that handcapping is a feat of skill, so it's actually possible for someone to beat it.

I'm not so sure handicapping is a feat of skill. A good season or two might just be luck. Maybe nobody picks enough games to show true skill.

How do you explain that if you bet the home underdog in NFL on every game from 1983 to 2007 you would have won 54.28% of the time. Enough to be profitable and no skill involved. Or if you just bet every underdog between those same years, you'd have won 51.5% of the time. Enough to be profitable. (Source - Wizard's Sports Betting Page)

ZCore13

To put it briefly, there is no great service for sports picks, and if there was, it would become ineffective very quickly.