Betting CFL Totals (Read if you Like Data)

Note:  Pointspreads.ca's in-house CFL handicapper Nick Christenson shared the following write up about totals.  Check it out, it's a great read if you like betting Over/Unders:

In the first couple of weeks of the season, CFL scoring was way down compared to where we’ve seen it in previous seasons. Then it started to pick up, but stayed the lowest we’ve seen in recent history.

Now, in recent weeks, totals have actually been trending down again, something that I didn’t expect. The average YTD game total is 43.27. The average game total in the last four weeks is 43.25.

Has scoring peaked for the season? Does this represent what we should expect from scoring over the balance of the year?

It may be tiresome, but I’d be remiss if I didn’t point out one again that these are extremely small sample sizes. Average total in the NFL varies significantly week to week, and it takes four CFL weeks to provide the same number of data points as we get from the NFL in one. There will be about as many CFL games all season as we will get in one busy week in the FBS division of American college football. Nonetheless, we need a baseline hypothesis, and while I’ve been a believer in the “return to recent norms” theory so far this year, I no longer think that’s the most likely model for predicting totals going forward, although it’s still entirely plausible.

Discarding the first two weeks in a season with no pre-season, the per game average total for weeks 3 through 12 has been 44.74. The opening totals for this week average 44.375. This week’s bye team, Winnipeg, with what is arguably a historically good defense, has been second lowest totaling team in the league this year at 39 PPG. Consequently, if you think that weeks 3 through 12 represent a good baseline, which is a part of the model I’m currently using, then an average of 44.375 is about a half point or so below where it probably should be, but it doesn’t leave a lot of room for betting games blind. This is factored in to my totals predictions, though, so if you believe in a different baseline, you can use that to adjust my totals numbers accordingly.

Quarterbacks continue to be on the move. This week it was Toronto dealing Nick Arubckle to Edmonton, who found themselves with a roster spot after dealing their once franchise QB away for pocket change. I don’t expect Arbuckle to play, or even be on the Elks active roster this week. I do expect it will be Taylor Cornelius, who is 0 – 3 starting for the Elks, under center in the late game on Friday. Even though Montreal acquired Trevor Harris from the Elks, given how he’s played, I expect this is Matthew Shiltz’s job at least until he either loses a game or Vernon Adams returns. I fully expect Shiltz to get the start this weekend. I haven’t heard who will start for Ottawa, whether it will be Caleb Evans again or former NFL backup Devlin Hodges. Neither looked particularly good last week, but their performance shouldn’t get more than their fair share of the blame for how the RedBlacks have done this year. Hodges may have more eventual upside, but he’s very new to the CFL game. Right now I don’t see any reason to adjust my Ottawa power rating based on news about who starts at quarterback.

Home field advantage got a boost after last week, and now stands at 1.65 PPG. If we look at how much the closing line values HFA this year (which is calculated by taking the sum of all the visiting team closing lines and dividing by the number of games played), the market thinks the home team in the games played so far should be about 2.3 PPG better than the road teams (your numbers may vary a little depending on whose closing lines you’re using.).  It’s not like all the best teams have been playing a road-heavy schedule and vice versa with the poor teams this year, so this doesn’t seem biased by schedule to me.

At the same time, HFA in the NFL has been 1.25 PPG this year with the closing line expecting a value of 1.87 PPG. so maybe home field in the COVID era really isn’t worth as much as it was for some reason. I still think that random fluctuation is the most likely explanation for this, but it’s worth keeping an eye on this to see how it develops.

Nick will return on Friday afternoon with this week's analysis.