Showing posts with label World Cup Betting Value. Show all posts
Showing posts with label World Cup Betting Value. Show all posts

Monday, June 14, 2010

One Metric to Determine Best World Cup Betting Value

By: T.R. Slyder,, @AndyDisco on Twitter

I combined's aggregator for the (betting) odds of each team to win the 2010 World Cup with Nate Silver's world cup (winning) odds to arrive at, given these two sets of data, the dollar amount that a person can "expect" to win with each bet.

What the numbers really tell us is that, for Spain, if it wins you would win $450 on a $100 bet (100 X 3.5 odds + the 100 back that you bet = 450). In combination with Silver's win percentage, that means that you have a 14.85% chance of winning $450. When you do the math, you get an expected return of $66.83 on a $100 bet.

In a sense it's like in Fletch when Chevy Chase says something like, "there's a 50/50 chance of survival. But there's only a 10% chance of that happening.".

So here's the breakdown. I went with only the top 9 teams in descending order of betting odds favoritism. I show their odds, their win payout, then the win %age as predicted by Nate Silver

Country- Gambling odds= (Payout if they win)- Silver's Win %age = Payout * %age

Spain- 3.5/1= ($450) 14.85%= $66.83
Brazil- 4.5/1= ($550) 22.39% = $123.15
Argentina- 6/1 ($700) 10.35%= $72.45
Holland- 8/1 ($900) 10.03%= $90.27
England- 8/1 ($900) 8.18% = $73.62
Germany- 9/1 ($1,000) 4.11% = $41.10
Italy- 16/1 ($1,700) 1.86%= $31.62
France- 25/1 ($2,600) 1.51%= $39.26
Portugal- 25/1 ($2,600) 3.54% $92.04

The single biggest explanation for this is that Nate Silver gives Brazil a huge relative chance (22.39%) of winning the World Cup, over the next likely candidate, Spain at 14.85%- or about 33% more likely. So by Silver's metric, Brazil is a clear favorite, but the betting odds have them listed as a slight second favorite.

We all know that a chain is only as strong as it's weakest link and, some of Silver's win percentages seem questionable (is Brazil really 5 times more likely to win than Germany? or .33 more likely than Spain?), but again, I wasn't claiming to personally swear by all of this data, I'm just saying, "Hey, I did this and thought you might be interested.".

A Few Clarifications for Gambling Rookies

You may be asking what the difference is between betting odds and win odds, since after all, the betting odds are the odds to win it all. The difference is that Silver's win odds, are (his interpretation of) the objective odds to actually win it. The betting odds at various betting sites, are generally an accurate reflection of the probability of winning, but are ultimately designed to entice betting. They are running a business after all.

To help illustrate the difference between the two, consider a hypothetical: If me, you and 9 of our friends fielded a team in the world cup, we would have a 0% of winning (true odds), but the odds-makers would offer a bet on us anyway, just to entice betting, thereby making them money. They would give us odds of 750,000/1 (gambling odds), and gladly pocket the money that our friends bet on us, or people accidentally/drunkenly wager on us, knowing full well the odds of us winning is zero.

Now take that theoretical into a real life example. Instead of you, men and 9 friends, imagine you are in England selling bets to the English betting public. They know damn well English fans WANT to bet on England, regardless of the odds, thereby making it a seller's market. So if England really has a 10/1 chance to win, the oddsmakers want to sell odds at 8/1, since it's a seller's market. And if England does win, that will save them 20% of all payouts. However, the betting public can also shop around a little and take whoever gives the highest odds. This may give our sportsbook incentive to drop their odds to 9/1 in order to offer more value.

Most of the oddsmakers represented at are from England, so in this exercise, I'm not sure how close to true odds these sports books are because so much of their betting is hinged upon taking money on England.

Note: All of the odds used were current at the time of writing.

That's how I roll.