New York Rangers: Understanding the team through analytics

Like any other area in which there is competition, professional hockey is constantly pushing the envelope to discover market inefficiencies. Through analytics and advances stats, the New York Rangers can paint a clearer image for projection.

The sports world was given a new perspective on analyzing team management on a grand scale following the 2003 release of Michael Lewis’ “Moneyball,” an inside look at how the Oakland Athletics had built a competitive team through the use of advanced statistics. Instead of putting an emphasis on buying marquee players, the A’s front office put an emphasis on getting on base to try and score as many runs as possible.

In the hockey world, pretty much every NHL franchise has a department of analytics in some shape or form. In some cases it is just one or two people, the New York Rangers’ director of Player care & development/analytics and hockey technology is Jim Sullivan. Although the Rangers have a lot of titles rolled into one with Sullivan, it does not mean analytics are not valued.

In relative terms, the strides the analytics community has made in the firmly entrenched hockey world is nothing short of remarkable. As some franchise’s stubbornly cling to the past, there are teams exploiting market oversight to assemble under market value cost teams. Finding cheap supporting cast players is the difference between being the Pittsburgh Penguins and doling out multiyear contracts to bottom six forwards like the Vancouver Canucks.

The New York Rangers were an objectively terrible team last season, using analytics can paint a crystal clear image as to why the team struggled so badly. All stats below were collected from hockeyreference.com

Control of the puck

The basic idea behind both Corsi and Fenwick is more shots on goal likely means more goals. In theory, if a team is generating more shots than the other team, the team is controlling the puck and therefore cannot give up a goal. Corsi is a measure of shot attempt differential while at even strength. The baseline for this stat is 50%, if a team has a Corsi of over 50% they are a positive possession team because they have generated more than half of the shots in the game.

Fenwick is a similar stat that measures shot attempts that are shots on goal or missed shots, blocked shots do not count. Similarly to Corsi, Fenwick’s baseline is 50%, if a team is creating more than half of the shot attempts they have a positive Fenwick.

In addition to these baseline stats of Corsi and Fenwick, there are also variations of these stats known as Corsi and Fenwick relative respectively. The relative stats are a measure of a singular player’s Corsi and Fenwick relative to the rest of their team. For example, Ranger’s forward Chris Kreider had a Corsi relative of 6.9%. This means that Kreider’s Corsi percentage of 51.2% was 6.9% better than the Rangers as a team.

Understanding the Ranger’s struggles

It is no secret that the Ranger’s struggles this past season stem from their blue line. The team’s collective defense was an abomination that failed to transition to offense from defense with any measure of success. Following the Ryan McDonagh deal at the trade deadline, the team lacked a true number one defenseman that could play at a positive possession measure.

The Rangers defense for the final two months of the season consisted of Brady Skjei, Neal Pionk, Marc Staal, Rob O’Gara, John Gilmour, Steve Kampfer and Ryan Sproul. This group was arguably the weakest in the entire NHL during the final stretch of the season. Using Corsi, it is clear that the Ranger’s defense was hemorrhaging shots on a nightly basis.

Of these seven defenseman, four, Staal (43.1 CF%), Neal Pionk (41.2 CF%), O’Gara (42%), and Kampfer (42.6 CF%) were negative relative to the rest of the team. Meaning that whenever any of these four defenseman were on the ice, they were dragging down the forwards because they did not generate shots. In the modern NHL, defenseman cannot afford to be one dimensional and be a liability in the offensive end. It would be one thing if any of those four defenseman were elite shot suppressors and lacked an offensive component.

Pionk did manage to post 14 points in 28 games as a rookie which should be considered encouraging, however he needs to round out his game to find a permanent place in the Ranger’s future.

The Rangers will likely improve in this measure because it would be pretty much impossible to get any worse. Add in the return of a healthy Kevin Shattenkirk and the defense should not be a total liability. Analytics give a greater level of depth to understanding how a team operates in a way that the eye test cannot. Although analytics are not the entire picture, they are a part of the story that should be used in talent evaluation.