NHL Early Favourites

At the start of last season, I read that NHL favourites are undervalued, "especially early in the season". 

Such pronouncements should always be verified before being acted upon of course, and even if you are happy with a strategy, markets change and they can often be improved upon. 2005 was also a long time ago.

Going into last season, my NHL strategy going back to the 2013 season (2012 was shortened due to a lockout) had a record of:

Backing all favourites over the same four seasons had a negative ROI of -0.6% and in addition would have required placing 5,183 bets, not a practical number for most of us. Looking at all favourites since the 2005 season and I came up with an ROI of -1.8%.

The 153.93 units was also meaningless since not everyone calculates returns based on a level one point stake.

As for the 'early in the season' part of the assertion, while 'early' is a relative term and could mean anything from the first week to the All-Star break, the results from backing October favourites were promising, an ROI of 6.8% from 619 bets, or five a day.

Not an improvement on my own system, but one to watch perhaps. In the event, October 2017 saw losses of 7%, and this month is currently down 10.52 points with a third of the month to go.

All favourites in 2017 had a losing ROI of 0.3%, while my record since 2016 is:

This season's one loss was on Sports Equinox Thursday - clearly the pressure of such a rare occasion was too much. Selection number six play tonight.

The same web site suggested that a subset of favourites, specifically "rested favourites on the road" were also undervalued.

True or false?, I hear you ask.

Again looking at the four seasons 2013-2016 I found that while backing all road favourites had a positive return of 3.5%, playing only rested teams increased this to 8.3%.

Again, not an improvement on my own system, but in 2017 this rested road subset continued to be profitable (6.8%) although so far is a losing bet this season.