clock menu more-arrow no yes

Filed under:

Measuring Lightning Players by Goals Above Replacement (GAR)

New, comments

A quick look at the Lightning using the new GAR metric.

NHL: Tampa Bay Lightning at Pittsburgh Penguins Charles LeClaire-USA TODAY Sports

Last week, @DTMAboutHeart published a series introducing his Goals Above Replacement (GAR) metric . If you enjoy reading about numbers and analytics, I can’t recommend strongly enough that you read his work at He’s done an excellent job of taking what could be a complicated topic and explaining it in an accessible way. He also provides links to previous work so that anyone can understand his methodology and thought processes. The series is presented in five parts that began on Monday and ended today.

Just in case you don’t have time to read the background work, I’ll do my best to provide a brief overview before diving into the Lightning’s numbers. If you follow baseball, you’ve probably heard of Wins Above Replacement (WAR). DTM has used that basic approach to create a similar statistic for hockey.

GAR shows how many additional goals per sixty minutes of ice time that a player accounts for compared to a replacement level player. A replacement level player would be defined as one whose production is at the bottom of NHL caliber players because that player could be easily replaced either with another low end NHLer or even an AHLer. By definition, a replacement level player would have a GAR of 0. We would expect every NHL player to have a positive GAR because a negative GAR indicates that the player is below replacement level and could thus be replaced with a better player fairly easily and with little cost. The most common exception to this basic rule would be for young players who are still developing. If you read DTM’s work, you can learn more about this definition.

DTM’s approach separates offensive and defensive impact so that we can see GAR for both parts of the game. For context, Brendan Gallagher had the highest offensive score last season at 0.660 GAR per 60 minutes, Chris Tanev had the highest defensive score at .20 GAR, and Crosby had the highest overall score at .87 GAR. Now that we have some context, let’s look at how the Lightning stack up by these measures.

First, let’s look at the best seasons by Lightning players going back to 08-09, which is as far back as we have the data needed for DTM to compute GAR. The list contains all the players you would expect to see as having the best seasons in this time frame. Martin St. Louis, Vinny Lecavalier, Steven Stamkos, and Nikita Kucherov are all among the Lightning’s best forwards in franchise history. Based on these numbers, one could make an argument that not only are Victor Hedman and Anton Stralman possibly the best defensive pair in the NHL right now but they are also the two best defenders to play for the Lightning at any point in the last nine seasons.

Now that we have some historical context for the Lighting, let’s look at last season’s numbers. Players far to the right on this graph excel offensively and players far to the top excel defensively. The gray area in the background represents all of the NHL skaters who played enough minutes to qualify for DTM’s calculations. Anton Stralman stands out as one of the best defensive players in the league last season and Nikita Kucherov looks to be not only one of the best offensive players in the league but one of the best players overall.

Finally, let’s look at each Lightning player using last season’s numbers. These are simple bar graphs, but for context, I’ve added the player’s percentile ranking in each measure for players at his position. A player with a percentile ranking near 100 is one of the best players in the league by that statistic and a player with a percentile ranking near 0 is one of the worst.

Here we can see just how good Nikita Kucherov is and how solid he is at both ends as well as how great the Hedman-Stralman pairing is. The numbers confirm the idea that part of what makes the Swedefenders so great is that Hedman excels offensively and Stralman excels defensively, resulting in a dominant complimentary pairing.

Steven Stamkos, by contrast had a good offensive season but a poor defensive season posting a negative defensive GAR that ranks in the 1st percentile among qualifying players. For context, last season was Stamkos’ worst by GAR and his first year with negative defensive GAR since 09-10. While it’s somewhat concerning to see him ranked this low, I would guess that last season will be an outlier amongst his prime seasons and he will rebound with a better season in 16-17.

Two players of particular interest are Jonathan Marchessault and Vlad Namestnikov. Vladdy fits right in line with other Lightning forwards like Ondrej Palat and Tyler Johnson, indicating that he is more than capable of filling his current role in the Bolts top six. His defensive excellence is particularly noteworthy. Marchessauly, as everyone by now is aware, left for Florida and is proving that he was deserving of a top six role.

New metrics like GAR are important in helping us better understand the game. While a single metric will never tell us everything we need to know, it can help us get a quick overview of a player and identify areas to dig deeper to try and learn more.

If you have any specific requests that you would like to know about how Lightning (or other teams’) players have scored by this measure, feel free to ask in the comments or @ me on Twitter and I’ll be happy to grab it for you.