Ramblings: Perfetti and Hintz sign; evaluating shot models; Klefbom; Pageau; Strome – November 10

Michael Clifford


The Jets signed their first-round pick, Cole Perfetti, to a three-year ELC with an AAV of about $1.6M. Perfetti is hopefully the 2C of the future they'll need to play behind Mark Scheifele.

Despite a lot of talk around Quinton Byfield, Marco Rossi, and Tim Stuetzle, let's remember that if any of them step into a 2C role in 2020-21, Perfetti will have the most to play with at even strength. Because of the timing of the 2020-21 season, I think we could see more rookies in the NHL than in prior years. I guess it could matter how their development this fall/winter goes. I'm not expecting Perfetti in the NHL from Perfetti this season but I am going to be keeping a very close eye on training camps.


If you haven't yet, don't forget to pick up your copy of the 2020 Dobber Hockey Fantasy Guide!


A thought-provoking read from a member of the online hockey analytics community, Draglikepull. For years, hockey analysts have been arguing about the value of shots. Just how valuable and predictive are shots, shot attempts, goals, scoring chances etc? Well, as DLP notes in his research: rely on scoring chances.

I very much encourage readers to go through his piece. It is not long, the math-y parts are clearly explained, and the implications could be wide-ranging for fantasy hockey. My projections (and the projections of many others at this site and others) rely, more or less, on the same data. It is just how that data is treated that makes the difference. Knowing where to look, and where not to look, is vital to being able to predict the future with any sort of precision.

We talked at the beginning of the 2019-20 regular season about irregularities in shot location data. Now, scoring chance data relies on shot locations like expected goals, just not for the same reasons. It really is fascinating that DLP's work in this regard, as he mentions, replicates work he has done in the past. As he mentions, this is not the first time that simple scoring chances have come out of this type of work looking better than more advanced models.

At this time, just assuming we should use scoring chances and only scoring chances is probably a bit premature. There is more work that needs to be done (and I'm not nearly smart enough to try and replicate his work). But it is fascinating, nonetheless.


Roope Hintz has re-signed with Dallas:



That leaves Dallas with almost no cap space, so their off-season is finished unless there is a trade in the works.