clock menu more-arrow no yes mobile

Filed under:

NHL teams would rather pay for points than wins

Points are nice, but wins are better.

If you buy something from an SB Nation link, Vox Media may earn a commission. See our ethics statement.

Tampa Bay Lightning v Toronto Maple Leafs Photo by Claus Andersen/Getty Images

Last week, Evan Rodrigues of the Buffalo Sabres received an arbitration award of $2 million for next season from an arbitrator. The team offered a one year contract worth $1.5 million while Rodrigues requested $2.65 million. On the surface, this seems like an inconsequential negotiation between a mediocre team and one of its depth forwards. But the submitted salaries and the award in this case are indicative of a larger issue in the NHL that will become more prevalent and more discussed over the next few summers.


After years of hard work by analysts with public data, hockey is starting to embrace new methods of player evaluation. While the loudest voices are often those screeching at either opposing end of the spectrum, many fans and most people close to the sport are somewhere in the middle of the process of learning and adopting new tools to help them analyze the game.

Shayna Goldman has been tracking the growth of research and development staffs in NHL front offices and almost every team has publicly acknowledged they have at least one full time person working on quantitative analysis. With the influx of new data coming in the near future via puck and player tracking, those staffs will grow as smart teams try to get ahead of the curve and not so smart teams play follow the leaders. Even five years ago, this type of investment from NHL teams was uncommon. Teams are embracing data. They might not be doing it quickly enough for a partisan like me, but they are doing it.

Because of that, we’re starting to see a change in the way teams evaluate players. More teams are focusing on speed and skill. They’re limiting the minutes of players who offer only physicality for physicality’s sake. Coaches who don’t adapt to the new style of play are relegated to the last outposts of old time hockey and eventually fading out of the league.

These changes impact players, as well. Tyler Myers signed a big contract this offseason, but from the reporting, he never seemed to have much of a market beyond Vancouver. Five years ago, a player his size who played as many minutes as he did in Winnipeg would have sparked a bidding war. The profile of player most coveted by NHL front offices is changing, and with that change in player evaluation comes a change in how players are compensated.

Or at least, it should.

The current contract market

Traditionally, player contracts have mostly been based on points and ice time. Skaters who play lots of minutes and put up points get paid with points being more important for forwards and ice time being more important for defenders. With new metrics available to teams in negotiations, expect that to start changing. But has it started to change yet?

We can look at this question in a couple of ways. The first, and most direcy, is to measure the relationship between points and contract value and compare that to the relationship between WAR and contract value. In this example, we’re using WAR as the new metric as it does the best job of capturing overall player impact in a single statistic. The plots below contains all contracts signed so far this summer as of July 31st.

The WAR stat used here and all the contract data is via Evolving Hockey.

The visuals tell the story but we can put numbers to it as well. Points have a stronger relationship with contract value than WAR. The correlation value (Pearson’s r) for points is 0.80 and for WAR 0.59, meaning the strength of the correlation between points and contract value is about a third greater than the correlation between WAR and contract value.

That shouldn’t be surprising. While new metrics like WAR are slowly gaining acceptance, they’re not yet prevalent enough to change years of contract negotiation and compensation history.

So what does any of this have to do with Evan Rodrigues? Well, Rodrigues is the type of player most impacted by the current process for determining contract value.

A few months ago, I wrote about the concept of empty calorie scoring, which is an idea prevalent in hockey for as long as I can remember that posits certain players score lots of points without doing much to help their teams win. To summarize the article, I found that those players do exist but they don’t always fit the popular narrative in the media.

But what I also found was that the opposite type of player exists: one who contributes to winning without scoring many points. And that’s Evan Rodrigues. Last season, he was in the 80th percentile among forwards for WAR with 1.7 but just the 51st percentile for points with 29. Given that we noted above player contracts are heavily based on points, that suggests he will be undervalued by his team. That proves true based on the arbitration numbers listed at the beginning of the article.

Rodrigues’ case is an interesting anecdote, but one player doesn’t tell us much. To get a better picture, the following chart shows the relationship between the degree to which a player is an empty calorie scorer with their contract value. To read this, a player far to the left would be an empty calorie scorer and a player far to the right would be a player who contributes to winning without scoring much.

The trend here is weak but noticeable. The correlation is about -0.31. Teams pay more for empty calorie scorers than they do for players who contribute to winning without scoring. That’s a market inefficiency. We shouldn’t see any trend here at all. It shouldn’t matter how a player contributes to winning, whether by scoring or not. They should get paid based on how much better they make the team, both because that’s fair to the player and because that’s optimal for the team.

But that’s not the case. For all the hullabaloo in hockey about valuing players who know how to win, NHL front offices overpay for scoring and undervalue the rest of things that players do to help win games.

The reasons for this are many. The biggest is momentum. For so long, contract negotiations between teams and agents have been based on points, time on ice, and even plus minus. Those conversations carry propulsive weight and repeat themselves every summer. The players are different, the numbers are slightly different, but the conversations are the same. Everyone knows the data points that form the framework for the discussion.

How contracts will change in the future

Change is always hard. If a team tells an agent that a 60 point winger is replacement level, the agent will be incredulous. Vice versa occurs if the agent tells the team a center with 30 points deserves to be paid like a second liner. If they deconstruct the framework, negotiations become riskier. The comparables that served as guides are no longer applicable and both sides are adrift in a nebulous space without clear markers, exposed to risk. And neither side likes risk.

So these changes, which are inevitable, will happen more slowly. Rodrigues will not be the test case. He’s not high profile enough. Slowly, teams and agents will begin incorporating new metrics, feeling out the process for how to adjust player valuations without tipping the whole boat. The CBA states that teams and agents can present information on the “overall contribution of the player to the competitive success of failure of his Club in the preceding season,” which opens the door to new metrics. Both sides would be negligent to not begin figuring out how to use them.

Then, at some point in the next few summers, a player with a big gap between their traditional statistics and their modern statistics will become a flash point. A team or an agent will take a stand based on better data, and we’ll have this discussion more broadly. Everyone will weigh in, even Pierre McGuire. It will be exhausting but necessary. Then, we’ll have it 10 more times for 10 more players. Those conversations will begin to slow as the idea of incorporating new metrics as teams and agencies develop them becomes common instead of a sudden shock.

But for now, teams are often paying for the wrong things. And players who provide the right things aren’t being compensated to the level they deserve. Some teams are probably aware of that and will be looking to make changes to better align their spending with player performance. Some teams are still blissfully unaware, happy to pay for points and ice time regardless of what else is happening when that player is on the ice.

The impact on the Lightning

The Rodrigues case is an interesting one for all the reasons listed above, but it should be of particular interest to Lightning fans. The Bolts have two players that fall into this category who will be up for be contracts next summer. Anthony Cirelli and Erik Cernak are both players who, so far in their young careers, have shown an ability to post high end WAR totals without scoring many points. If they continue that next season, they could be next in line to sign a deal in a market that undervalues their skill set.

For the Lightning, that could mean the ability to sign impactful players at a discount compared to their true value. For the players and their agents, that means deciding how hard they’re willing to push against the traditional valuation process to get the best possible contract.

Cases like this will continue to come up every summer. So far, we haven’t seen much of a tide change. But it’s coming. And the Lightning could be on the front end of it next summer.