Introduction to Comparables

A brief note on disappearing: this is my first article in some time. It kind of just happened. It's not that I had totally lost interest in hockey. I watched some games from start to finish this year, and watched the Oilers at least little bit too, but I didn't watch a full game involving the Oilers from start to finish all season. This was very, very good for me. A self-imposed season-long exile had me feeling much less frustrated and cynical this past April than I was at the same time a year ago, and some of the more recent developments have me feeling downright excited! Given that I've always liked the draft, which is only a week away, this seemed like a natural place to step back in.

There's already a lot of great information available on the draft: the major scouting services are all providing regular updates, the SBN mock draft is in full swing, and Bob McKenzie's final draft list is already online. One of the ways I've tried to differentiate myself from most of this work is by looking at statistical comparables. I did one for Leon Draisaitl after the Oilers drafted him last year, and was more actively involved doing a bunch of them in the leadup to the 2013 draft. With a week to go before the draft, this will fall somewhere in between. The plan is to do two per day for six days starting tomorrow, and I'll be asking for reader input on who to profile.

The methodology is quite simple. I'll be considering the league each guy played in (note: the CHL players will be compared to one another across leagues), whether or not the player is in his first year of draft eligibility, where each player is projected to go in the draft (based on McKenzie's list), and the amount of offense each player brings to the table based on his performance during his draft year in the regular season and playoffs. Since I'll be looking at players drafted as far back as 1981, some adjustments will be necessary. As I'm sure you all know, teams scored a lot more goals in the 80's and early 90's than they do today. The graph below shows how many goals per game the average team scored in the three CHL leagues since 1980-81:

In order to take the historical scoring rate into account, I will adjust each player's offense to a league that scored 6.8 goals per game. For example, Ryan Smyth was drafted in 1994 after scoring 0.69 goals per game and 1.46 points per game; but Smyth played in a league that scored 8.44 goals per game, which is 19.4% more than the benchmark, so his totals are adjusted down 19.4% to 0.56 and 1.17 respectively. I think it's a pretty reasonable way of doing things, but I'm open to suggestions.

Of course, it's always important to remember that each player is unique. The players with the best offense generally do well, and those with the worst generally don't become stars, but there are exceptions, especially at the top end. Sometimes a player lights things up in the CHL but absolutely craters once he gets to pro hockey. Looking at the fifteen best and fifteen worst CHL forwards drafted in the top ten in terms of adjusted points per game since 1981 demonstrates both points well:

The first list is filled with stars plus the occasional bust to remind us that no one is sure thing. The second list is populated with busts, role players, and Ron Francis, reminding us that teams really didn't do a very good job of drafting in the 80s and 90s (special note: if either Pavel Zacha or Lawson Crouse is drafted in the top ten, he'll become the first player to take a spot on the second list in fifteen years). It's not a perfect measure, but CHL offense is a pretty good rule of thumb.

Tomorrow's comparables: Connor McDavid and Jack Eichel