Saturday, February 10, 2018

Buying NHL Face-off Wins

What does it cost to buy Face-off wins on the NHL free agent market? If you're looking for a guy to take face-offs, that most likely means you are looking for a center. The players analyzed will be those who are pending free agents, so we can see the dollar figure they commanded on the open market. Centers tend to cost more on average than wingers, given their importance to team success.

For a better look at the free agent forward market (for forwards at least 26 years old) and how many Face-offs they average per game (from 2005 to 2016), there is a histogram below (the sample contains 936 players). This shows what proportion of the FA forward market take Face-offs and how rare the high-volume guys can be.

58% of the sample will average under 2.4 Face-offs per game, with 66% averaging less than 5. So we are dealing with a small proportion of free agent forwards who are even capable of taking a significant number of Face-offs per game, and an even smaller number of those who are capable of winning over 52% of their draws. Finding a good Face-off man on the free agent market can be difficult, especially for one capable of also playing a top 2 line role.

Expected Salary by Face-off wins

What exactly can NHL teams expect to pay for FWs on the open market? We are accounting for salary cap inflation with the variable E[2017 AAV] which is what would the AAV have been with a $72M cap.

If you want to buy a center who can average over 10 FWs per game, it will cost you. This is generally because the highest volume guys are the top liners who play in all situations at both ends of the ice. The cheapest Center to average over 10 wins per game was Manny Malhotra in 2009/10 who signed a 1 year $700K contract with San Jose. Manny was coming off a $1.2M AAV salary, then followed it up with a $2.5M salary, so that one-year cheap deal was an anomaly.

Below is a scatter plot charting Face-off wins against salary for this same population of free agent Forwards. It does not show the term of the contracts, only the salary AAV. This sample is different than the ones above, in that it caps the age of the pending free agent at 32. The three yellow numbers represent groupings of players that are discussed below.

It is hard to fit any Excel trendline to this data, given how the forwards who don’t take many Face-offs affect the sample. The Top paid wingers (Group 1 above) drag the line up for the low totals. We get a negative slope into the population of low-end Centers, then it starts to climb into the high-volume Centers. Group 2 above are the skilled Centers who tend to share ice time with other centers (the win totals are not low because of bad win %), like Evgeni Malkin, Patrick Marleau, and Danny Briere.

Group 3 above are the discount Centers like Boyd Gordon, Zenon Konopka, David Steckel, Jay McClement, Kyle Brodziak, Matt Cullen, etc. One of the few places where you'll be able to find good value on FWs, is for veteran depth centers (Group 3). They average less ice time, tend to kill penalties, and take a high number of defensive zone draws. There are a few of them available every summer. Within this group, there was a 0% correlation between Face-off winning % and salary. If you're good enough to still be taking NHL Face-offs over the age of 30, then teams aren't overly concerned about the winning %.

Expected Salary By Face-off Winning %

If we look at the winning % of a large sample of pending Free Agent Forwards who took at least 100 draws and what they earned on the FA market, we see a scatter plot with almost no correlation. Ergo, all you need to do is be someone who takes Face-offs to get signed, there is not a significant wage premium for those with a higher winning %.

No comments:

Post a Comment