I’m not anti-analytics (although I think people go a little overboard on Corsi) but this is the problem when you rely on them too much. Almost everyone agrees that coming up with a WAR model is more difficult for hockey than any other sport and there are at this point about 5 competing yet flawed models. So when someone on Twitter says player x is soooo much better than player y because look at his WAR number the first thing someone should ask is, which WAR model and do the biases of that model influence this particular comparison? But no one ever does. Statistical models, unless they have a very high r-square value are all subject to the biases of the people who create them. WAR is a model made up of other models so the problem is even worse.

Cale Makar doesn’t have to make up for Barrie’s point total, Kadri does. And he almost certainly will. He’ll probably exceed it. With Barrie’s production shifted to the second line, anything Makar does is additive. If you don’t want to look at it that way then the other way to look at is Kadri plus Makar need to exceed Barrie plus Kerfoot. That’s a pretty decent bet as well. Also, it is highly likely that Burakovsky playing next to either Kadri or MacKinnon will out-produce every previous point total of his career. Not certain but likely. People can figure all of this out about the Avs, that’s why they’re picked to finish first or second in the division by just about everyone. The biggest question mark is in goal where Tierney for some reason has them strongest.

As you guys said, every bet can go wrong but moving guys like Bourque, Andrighetto and Soderberg out and replacing them with guys like Burakovsky, Donskoi and Kadri is very likely to improve a team that whacked Calgary and took San Jose to seven games. Again – not certain but likely. I highly doubt they’re going to finish 6th in the Central. Tierney creates some interesting charts but I think he’s pretty far off base here and will end up creating ammunition for the anti analytics folks.