Burke firing improves Stanley Cup betting odds for Leafs

Nate Silver has theorized that consensus builds models for better forecasts, but even moreso, having a financial stake in the outcome will force people to build better forecasts. Sports odds, by proxy, have a lot to offer.

Basically, sports odds represent the consensus forecasts of people willing to spend money, and they do it well. For a while, I was under the impression that perhaps popular teams got way more favourable odds than less popular teams due to the sheer volume of fans betting on them, changing the line. A quick look at records of Dallas Cowboys games against the spread showed me that, if that was ever a market inefficiency to be exploited by the betting market, it already has been.

Of course, they can be terribly wrong. The betting public had a better opinion than I did of the Buffalo Sabres coming out of Terry Pegula’s offseason before the 2011-12 season. The odds had them fourth in the conference, behind Boston, Pittsburgh, and, er… Washington. Those odds also undervalued the Ottawa Senators, who vaulted into a playoff position with good coaching and an unexpected season from Erik Karlsson. 

Dramatic shifts in the betting lines are always worth noting. The Toronto Maple Leafs, as noted in the headline, have gone from 40:1 to 33:1 in the two days since Brian Burke has been fired. In layman’s parlance, that means a $10 bet will now get you $330 if the Leafs win the Stanley Cup as opposed to $400 when the line opened.

What changed? Well, perhaps the better markets believe that the Leafs are in a better position to win this season with a management change, which has been accepted by the public as the lubricant which will send Roberto Luongo to Toronto in exchange for whatever. Perhaps the betting markets have no confidence in Burke. Perhaps they know the ‘whatever’ isn’t as valuable for a single year run as Luongo. Perhaps it’s entirely coincidental.

However, from Silver’s book The Signal and the Noise: “Successful gamblers—and successful forecasters of any kind—do not think of the future in terms of no-lose bets, unimpeachable theories, and infinitely precise measurements. These are the illusions of the sucker, the sirens of his overconfidence. Succesful gamblers, instead, think of the future as speckles of probability, flickering upward and downward like a stock market ticker to every new jolt of information.” [1]

I’m not too up on sports betting, but I gather that 40:1 to 33:1 is a fairly dramatic shift over two days, even if 33:1 is still fairly long odds, and, since there are 30 teams, still considered to have less than an even chance at winning.

  • Stanley cup odds are not a balanced market because there’s no way to take the other side of the bet, so “public teams” will often have worse odds simply because square bettors aren’t going to be as sensitive to price movements.

    So Burke put the Leafs in the news, square bettors hit the line, and it moved. Nothing to see here.

  • No, the other side of going long is going short. If I get a point spread, I can bet the favourite or the dog. On the Stanley Cup odds I can either bet on the Leafs winning or….not bet. That’s not the same. There’s only buy pressure on one side. If I think the Leafs line is too expensive, how would I go about betting against it? I can’t.

    A better indicator is the season points line because you can bet both sides. And that didn’t move. So there’s your answer. Maybe Nate Silver will write about this in his next book so that you understand it.

  • It’s interesting to see that the odds have changed (although.. whose odds are these?), but I agree with mclea that it’s not very significant.

    Going from 40/1 to 33/1 is actually pretty much insignificant if you translate them to percentages.

    33/1 odds translates to 1/(1+odds) = 1/34 = 0.0294 = 2.94% chance of winning the Cup

    40/1 odds would be 1/41 = 2.44%

    So half a percent change, 0.50%… That’s like going from 13/1 to 12/1 (a 0.55% increase).

    You can get a sense of the accuracy of a set of odds by converting them to probabilities and adding them together. If the total is way more than 100%, then the odds are not accurate, there is a huge cut, and I think that cut (inaccuracy) is most noticeable in the low-probability teams.

    If you do this for the odds for all the teams, you’ll find the percentages add up to 130-150% (depending on your source of odds, anyway). I think the top teams have relatively accurate odds, but the bottom half of the teams are probably all inflated by around 1-2% (which is significant when the probability is in the low single-digits to begin with). So if the true probability for the leafs to win the cup is closer to 1.5%, then you should be getting
    odds = 65.67

    so about 66/1 (which still seems high to me)

    People betting that the Leafs *might* have an *increased* chance of getting Luongo and that that brings their odds up above the old, misleading 1-in-40… well I think they are throwing their money away.

    “Basically, sports odds represent the consensus forecasts of people willing to spend money, and they do it well.”

    Personally (and especially in a case like this when both sides of the bet aren’t allowed, as mclea pointed out), I think sports odds represent a consensus of people willing to throw money away 😉