What do the Betting Markets really say about the Leafs?
by Hawerchuk on Nov 19, 2009 9:15 AM EST
in Bold Predictions
Email
PrintOn Tuesday, I looked at some dire predictions for the Toronto Maple Leafs: they supposedly had just a 1.7% chance of making the playoffs, and a 1-in-125 chance of winning the Stanley Cup. I pulled up the Stanley Cup odds for the bottom seven teams in the league:
| TEAM | 10/1 |
11/18 |
| Tampa Bay | 100/1 | 75/1 |
| Florida | 30/1 | 75/1 |
| Nashville | 40/1 | 75/1 |
| Atlanta | 50/1 | 75/1 |
| Minnesota | 40/1 | 100/1 |
| Toronto | 60/1 | 125/1 |
| N.Y. Islanders | 100/1 | 150/1 |
The Leafs have played quite poorly, but their Stanley Cup odds have dropped because their odds of making the playoffs have dropped, not because they're perceived to be substantially worse than they were at the beginning of the season. Before the season started, the Leafs' upside appeared to be the 14th overall seed in the playoffs - last year, on the day the playoffs started, the 14th seed (Montreal) was given a 2.86% chance of winning the Cup.
If that 2.86% holds true for the Leafs should they make the playoffs, then the futures market gave Toronto a 58% chance of making the playoffs prior to the season and are still giving the Leafs a 28% chance of making the playoffs. (This implies that the Leafs are a .523 team, which is an overestimate.) This isn't a perfect futures market, and the people who set the lines hedge a bit so that they'll make money even if any unlikely outcome occurs (this hedge is known as the 'vig'.) If I remove the vig (as Tango suggests in the comments below), I get 39% and 17%, respectively. Are these gamblers nuts?! Didn't we just hear that the Leafs have a 1.7% chance of making the playoffs?
If I then back out these playoff probabilities by running a simulation of the entire season, this futures market projected the Leafs to be a .487 team at the beginning of October, and, after the first 18 dismal games, it projects the Leafs to be a .502 team for the rest of the season. In other words, people who have money riding on this question think the Leafs are basically just as good as they thought they were when the season started and their performance through 18 games (now 19) is in no way indicative of their true talent level.
If we assume the Leafs are exactly a .500 team, then they have a 13% chance of making the playoffs after last night's loss. But no matter how you slice these odds, the Leafs were - and still are - perceived to be a middle-of-the-pack team, not the worst in the league. They just have a long road back into playoff contention.
12 comments
|
0 recs |
Do you like this story?
Comments
You have seriously brought some much needed things like “proof” and “calculations” to the coverage of the Leafs. I’m still not going to start hoping again, since I’m sick of those hopes getting crushed, but you keep showing me that it’s not the end of the world (despite what the Toronto MSM say), and for that I am grateful.
"The only way out is in a body bag. Go Leafs Go." - Blinky
by Karina on Nov 19, 2009 11:57 AM EST reply actions
Can you redo the odds by backing out the vig?
If you add up all the odds, I presume that you get something like 1.58 Stanley Cup winners. Multiply the odds by some constant in order to get the right value. Then, redo your odds on that basis.
I pulled up the Stanley Cup odds for the bottom seven teams in the league
They don’t have Anaheim listed in that group?
http://www.battleofcali.com/
Anaheim’s odds of making the playoffs are higher now than when the season started! So despite the crummy start, the bettors think Anaheim’s a way better team than they did six weeks ago.
And am I reading that link correctly?
The Ducks at “Open” (presumably before the season) had 25-to-1 odds at winning the Stanley Cup, and now that they are at 15th in the west they are now have “Current” 20-to-1 odds?
Vegas likes the Ducks’ chances more now?
http://www.battleofcali.com/
Thanks, Gabe.
These are dark days to be a Leaf fan. We’ll not be making the playoffs this year, but it’s somewhat consoling to read stats that are actually legit.
I’ve got a question Gabe: I don’t study regression until next semester, so my knowledge of it is somewhat sketchy at best, but at what point does the “mean” that Toronto’s statistically supposed to be regressing to, well… regress to a lower value?
Another way to ask it is, at what point with regards to statistics is this not a .520 team playing like a .300 team, but a .300 team in and of itself? (to use a (hopefully) extremely low-end example).
If the Leafs are still at .300 after 40 games, then – knowing nothing else about them – we regress them to a .400 winning percentage for the rest of the season. Teams tend to be pulled very strongly towards a .500 record even if they play poorly for a lot of the season – and that even includes teams that trade away all of their veterans or expansion teams that were legitimately that bad.
Incidentally, “regression to the mean” is not the same thing as linear regression (which is what most people mean when they talk about “regression.”) “Regression to the mean” refers to the difference between what you observe in a small number of measurements and what you’ll expect to observe in a large number of measurements. For example, after each team has played only one game, you have a lot of teams at 1.000 and a lot of teams at 0. This one-game record has only the tiniest relationship with winning percentage over the rest of the season, so our best guess is that each of these teams will go .500 the rest of the way. In other words, we regress 100% to the mean of .500.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
