Your cart

An Analytical Overview Of The 2016 NHL Off-Season

Since the National Hockey League endorsed analytics, their enhanced statistics can be used in contract negotiations, specifically when dealing with arbitration. Now that analytics are becoming more widely accepted, it is interesting to note the analytical value of those traded during this past off-season and the free agent signings. Whether or not analytics were actually considered in a signing, looking at players from this perspective can show value where standard analysis may not.

Casual hockey fans—and some teams—may prefer not to go deeper into trades or free agent signings during the off-season, only analyzing the move through an eye-test. The eye-test is a standard analysis, while enhanced statistics can provide a different perspective on a player’s skills. So, while some may not be interested in digging past the surface level of signings and trades, the data is available as a tool for those who are. Looking at all of the information available together could help teams and fans alike anticipate how the team will perform during the next season in light the off-season transactions.

Trading players and signing free agents without considering analytics has had negative consequences in the past. For example, the Philadelphia Flyers re-signed defenseman Andrew MacDonald to a six year, $30 million contract after trading for him from the New York Islanders. His analytics though, indicated that MacDonald’s contract could prove to be problematic for the Flyers––which it ultimately did, leading the Flyers sending MacDonald down to the AHL for much of the 2015-2016 season.

Get The Latest Sports Tech News In Your Inbox!

The trade of P.K. Subban from the Montreal Canadiens to the Nashville Predators for Shea Weber certainly was analyzed by the analytics community. Much of the criticism was of how Montreal handled this trade, including their dismissal of analytics consultant Matt Pfeffer. Prior to Canadiens’ General Manager Marc Bergevin finalizing the trade, Pfeffer reported to management about the analytical implications from this trade––emphasizing that the Canadiens should not make this trade since Subban is a better player based on almost every analytic metric.   

Pfeffer noted a major difference between Weber and Subban’s even-strength goal differential––with Subban at 3.14 percent and Weber at 0.18 percent. “With his experience, you really need to only look at goal differential to measure his impact. You only need Corsi if you don’t have a large enough sample size to evaluate goals. My analysis of Shea Weber had very little to do with Corsi. It’s easy to hate on Corsi, but (Weber) is not a good goal differential guy either. He’s not pushing the needle in terms of how many goals the Nashville Predators score and get scored on when he’s on the ice. He’s good, he’s serviceable, but he doesn’t really push the needle on either side,” Pfeffer explained. Pfeffer further said, “An average NHLer is worth a heck of a lot and that’s what Shea Weber is.”

From an analytical perspective, Subban’s shot generation and shot suppression were superior to Weber’s. His playmaking was also more impressive than Weber’s––in fact, it was quantified that Subban made 600 more plays with the puck than Weber during 5-on-5 play, in ten fewer games. Additionally, Subban’s 4.86 shots per 60 and 4.0% shooting percentage last season is better than Weber’s 4.78 shots per 60 and 3.54% shooting percentage. When factoring in their defense partners—Roman Josi for Weber and Andrei Markov for Subban—Josi’s playmaking numbers certainly could have contributed to Weber’s success.

Not only did the Canadiens disagree with Pfeffer, but management chose not to renew Pfeffer’s contract. While Pfeffer’s views on the trade may not necessarily have directly influenced management’s decision, it is widely believed that it did.

This trade though, likely did not consider the analytics. It seems that this trade went much deeper than what the analytics could show. Even before the trade, there was speculation that Montreal wanted to be rid of Subban. Acquiring Weber seems like a win for Montreal, regardless of whether or not he actually is an upgrade from Subban. If their analytics were not convincing enough, look at their contracts: Subban is 27, Weber is 31. Subban’s contract is until 2022 with $58 million remaining, while Weber’s lasts until 2026, with $48 million due.

When looking at free agency signings and their inclusion of analytics, David Backes comes to mind. The 32 year-old former St. Louis Blues captain signed a five year, $30 million contract with the Boston Bruins. Backes has been commended for his shot blocking and physical style of play that can deliver hard hits. However, signing a 32 year-old player that plays this style of hockey for six years, at $6 million AAV is risky since it is not the most sustainable style of play. The “eye-test” often favors a player that has a grittier style of play, typically over-crediting the hits and blocked shots. Plus, when it comes to a heart and soul former captain like Backes, his intangibles are also over considered.

Although Backes had success with the St. Louis Blues in the past, this contract should be concerning based on his overall offensive performance. A comprehensive metric to evaluate a player’s offensive performance is passing data. By looking at shot assists, not only are the shots a player takes assessed, but how a player sets up his teammates. Ryan Stimson of Hockey-Graphs created the Passing Project and has studied passing and shooting data to formulate evaluations of players’ and team’s offensive production. Based on this data, Stimson noted how the Backes contract does not look at Backes’ production from enough of an analytical approach: “Backes ranked 219th out of 392 forwards with at least 100 minutes tracked last season. His Primary Shot Contributions/60 were wedged in between Mike Santorelli and Reid Boucher, two players you wouldn’t hand over that sort of contract to.” To compare, Mike Santorelli is still a free agent, whose previous contract was worth $875,000 AAV. Reid Boucher was re-signed by the New Jersey Devils for $715,000 AAV.

While Backes may be viewed as more valuable player based on his physical play, his contract most likely did not consider his offensive performance and age nearly as much as it should have.

Another example of a free agency signing that likely did not involve analytics was the signing of Troy Brouwer by the Calgary Flames for four years, at $4.5 million AAV. His grit and Stanley Cup experience as a member of the 2010 Stanley Cup Champion Chicago Blackhawks add to his value. Brouwer certainly had an impressive run with the St. Louis Blues in the 2016 Stanley Cup Playoffs. However, that cup run alone represents too small a sample of his entire to career to offer him such a lucrative deal.

Brouwer could be considered consistent over his career, but not quite dynamic––which should be expected signing a contract of this length and cap hit. But Brouwer’s inefficiencies could have been affected by his offensive zone starts. In fact, when looking at all forwards that played at least 500 minutes of 5v5 last season, Brouwer ranked 66th in the most difficult zone starts. Still, having more offensive zone starts does not guarantee that Brouwer will be as successful as his contract reflects. Based on Brouwer’s career enhanced statistics, it is unlikely that Brouwer will be the ideal top line player alongside Sean Monahan and Johnny Gaudreau––especially since at the age of 31, it is unlikely that Brouwer’s analytics will drastically improve over the next four years of his contract. Brouwer could be fairly effective for the Flames, especially since his zone starts will be better on the top line, but maybe too much emphasis was placed on his playoff success rather than his entire career.

On the other hand, the Florida Panthers, a team that overhauled their front office to be more analytically savvy, made off-season decisions that reflected their renovated front office. The Panthers traded defenseman Erik Gudbranson and a fifth-round pick to the Vancouver Canucks for center Jared McCann, a second-round pick, and a fourth-round pick. While Gudbranson was on the ice for the Panthers, the team did not generate a high number or shot, nor did Gudbranson suppress enough shot attempts against his team. The Canucks follow a more traditional approach that favors a more physical, stay-at-home defenseman. While a stay-at-home defenseman could be successful analytically like, like Niklas Hjalmarsson of the Chicago Blackhawks, the Panthers did not see that in Gudbranson and decided to trade for a young, two-way center with analytical upside.

The Panthers did replace Gudbranson though through free agency signings like Jason Demers. Demers was signed by the Panthers for five years, at $4.5 million AAV. His possession rates and shot rates made him an analytically favorable signing for the Panthers. The addition of Demers continued the Panthers’ impressive defensive free agency moves, which included signing Keith Yandle and re-signing Aaron Ekblad.  

Brandon Pirri’s $1.1 million contract with the New York Rangers is very cost effective. Pirri’s analytics demonstrate his talent, in the quality and the rate of his shots. A player’s shot quality is determined by looking at the difference in the expected goals (which is based on both the location of the shots and type of shots) and a player’s goal totals. Pirri is a young player, at twenty-five years old, potentially with a lot of offensive upside based on his career analytics.

The analytical value of a contract should be considered in both trades and signings of star players, like P.K. Subban, and bottom-six players, like Teddy Purcell––whose $1.6 million contract with the Los Angeles Kings is very favorable based on his overall offensive production. Players like Justin Williams, who signed a two year, $3.25 AAV contract with the Washington Capitals last off-season, could point to their analytical value to drive up the value in their contracts. In addition to his Stanley Cup experience, Williams has excellent possession numbers that expand his value. There are still many free agents though, that have untapped potential because teams do not consider the analytical value that is quite telling of a player’s talent––like Brad Boyes and Kris Versteeg, both of whom are assets based on their passing and shooting data. There is more to consider than a player’s analytical value, like their age and experience, but it should be a major consideration that is far too often overlooked.

← Back to SportTechie