value in betting NHL OT inc ML vs REG only +0.5??

[pitthockey]

Ok I may be way off on this one so correct me if I'm wrong. Since the rule changes in the NHL (start of 05-06 season til present), there have been 1502 games played. 339 of these went to OT, about 22.57%. Is there +EV in betting both the OT included ML and the half goal underdog? (Stole the calculations from Homer's site.)

ex: Carolina vs Washington
OT ML on CAR is -123
REG only WAS +0.5 -133

Using 1000 for the base:

CAR -123 bet $1000 to win $813.01
WAS+.5 -133 bet $1034.89 to win $778.12

EV -$221.88

If either team wins in regulation, or the dog wins in OT you lose $221.88.
If the favorite wins in OT you win $1591.13.

This needs to happen 12.2% of the time to break even.
I haven't went through all the games yet but I'd say this happens roughly 15% of the time.

EV = (win% * win) + ((1-win%) * loss)
EV = (.15 * 1591.13) + (.85 * 221.88)
EV = $427.26

Risk $221.88 to average return of $427.26

My numbers may be slightly skewed, and this may not be a large enough sample size, but if this is all accurate this is +EV, no?

[beetman]

A game between evenly matched teams should be more likely to go to OT than a game between two disparate teams.

[pitthockey]

Yeah I'm thinking ML's -140 or lower.

[136913691369]

[ QUOTE ]
A game between evenly matched teams should be more likely to go to OT than a game between two disparate teams.

[/ QUOTE ]

Also, lower scoring teams like Phoenix and New Jersey are more likely to go to OT than higher scoring teams like Buffalo and Carolina.

I have a similar system to the OP, but mine involves totals. It is +EV according to my data based on the 05-06 season (1238 games), but the actual results (small sample size of around 20 games) have not been to good so far. OP if you wanna share ideas, PM me

[beetman]

But by the same token, a -130 favorite is less likely to win in OT. Your 15% estimate assumes the favorite wins about 2/3rds of the time that the game goes to OT, and that's certainly not the case.

[pitthockey]

Yeah I need to look into this a little deeper. FWIW -123 ML to win in the example would convert to 55.16% probability. 55.16% * 22.57 = 12.45 which is still slightly above the BE%. But again, numbers may be skewed.

[beetman]

A -123 favorite isn't going to win in OT 55.16% of the time though. They're a -123 favorite at the start of the game when there's at least 60 minutes of hockey to be played, but in overtime, there's only 5 minutes plus a shootout.