A Behind the Scenes Look at the NHL Salary Arbitration Process

I was invited to take part in a mock NHL salary arbitration for the National Academy of Arbitrators’ annual convention (which was held in Vancouver this year) earlier this month.

It was a week long conference that featured speakers on a broad number of topics, and this included sports arbitration. I played the part of “co-counsel”  for the player/NHLPA during this mock salary arbitration, and on the side of the league and team – my opponent, if you will – was a guy you may have heard of – former Toronto/Vancouver/Anaheim GM Brian Burke.

It was a very interesting day and it was fascinating to see the process play out in real time. The arbitration was moderated by real NHL/MLB arbitrators, and several current and former NHL arbitrators were in the audience, as well.


Our Case

For the mock trial, Cam Neely was used as the player filing for arbitration. Since it was only a demonstration of the salary arbitration process, we assumed that the year was 1989 and Mr. Neely (24 at the time) took his team to arbitration (player-elected arbitration). This assumption made the case very unique and also challenging. 

We used the current CBA provisions, which lay out clearly what is admissible and inadmissible in the process. I was a bit surprised to see that anything pertaining to the NHL salary cap was inadmissible as evidence. The salary cap is a very relevant factor when assessing a player’s worth to his team (or on the open market). For example, a $1 million player under a $55 million cap has a much different value than a $1 million player under a $70 million cap. That bit of information didn’t make much sense to me.

Essentially, my overall argument as Neely’s agent was structured to highlight Neely’s overall contributions beyond just goals and assists. Because so much of what he did on the ice (intimidate, play with an edge) was hard to quantify, I had my work cut out for me (especially considering time on ice and hits weren’t even recorded back in 1989). It was also very interesting to compare different eras, as Neely played in a much higher scoring time in the NHL than we see today.

Here are the provisions for NHL salary arbitration:


I used Neely’s PIM as a measure of his physicality – PIM isn’t always the best indicator of toughness and grit, but in Neely’s case, his increasing PIM points to a player learning to play with an edge more consistently. His PIM per season, starting in 1985, rose from 126, to 143, to 175, to 190 by the end of the 1988-98 season.

Special qualities of public appeal also factor in to arbitration cases – Neely being a ticket seller and a fan favourite, for example, were both relevant factors in this case.

The Comparables

Under the arbitration process, player comparables help to establish a market value for a player, but they aren’t the only way to do so. It isn’t like MLB salary arbitration where the arbitrator picks either the player or the team contract request – instead, it is a judgement call based on the facts presented from either side.

I used CanLII (Canadian Legal Information Institute) to dig up a few past arbitration cases to get a feel for how the process played out. Some of the cases I looked at: 

Neely was a productive NHL player in 1989, but he had yet to emerge as the premier power forward in hockey. So we essentially had to pretend that we didn’t know how the next few years of his career would play out. And even though he was a consistent 35+ goal scorer, it was a different era back then.

In Neely’s platform year (PY – a common term used in arbitration), he scored 37 goals, which placed him 26th in goal scoring in the league at that time.

I introduced Scott Hartnell and Corey Perry as example of players who were compensated for more than their offensive production (tying this back to the overall theme of our argument as Neely’s agents). However, for relevant comparisons, players have to be of similar age and equal contract standing (you can’t compare a UFA to an RFA, for example). So we used Hartnell’s 2007 season as his PY and Perry’s 2008 season – but the contracts were irrelevant from an actual dollar perspective, as the contracts were signed so long ago. They were just used to show how teams in the NHL rewarded grit, toughness, and intimidation financially.

In Hartnell’s PY, for example, he scored only 22 goals and had 39 points. The below table shows the per-game production for Neely compared to a few of our player comps (I found out during the arbitration hearing that David Backes was inadmissable as a comparable as he signed his contract as a UFA – I should have double-checked CapGeek before using him – thankfully Mr. Burke took it easy on me…).

Neely comps

Milan Lucic was our primary comparable. Some excerpts from our argument:

Both are rugged wingers who intimidate with fighting, hitting, and size and strength on the wing. Neely is 24 right now, and Lucic was 24 when he inked his contract in the summer of 2012. Mr. Lucic was rewarded with a three year 18 million contract in 2012, with a modified no trade clause. No trade clauses give players to ability to veto or turn trades down, and it is a bargaining tool used by teams to pay players a lower salary in return for more stability.

Mr. Lucic was 50th in the NHL in goal scoring in his platform year (26 goals), and 53rd in points in that year – 61 points. $6 million per season is typically compensation given to top line stars or elite defensemen. However, Boston recognized how valuable his other attributes were, and they compensated him for it. No player nowadays combines toughness and skill as well as Cam Neely – Lucic is the closest, and he is still an inferior offensive player

In his platform year, Mr Neely also had 55 more PIM than Mr Lucic, 11 more goals, and three more assists, all in seven fewer games Mr Neely added seven goals in 10 playoff games, compared to none in seven games for Mr Lucic. Both Neely and Lucic are proven playoff performers, and Boston recognized that with Mr Lucic’s contract. In 62 playoff games, Mr Lucic recorded 35 points.

In 44 playoff games, Neely recorded 34 points. Mr Neely is more experienced (419 games compared to 359), a better offensive player (166 goals to 90 and 320 points to 212), and a much more consistent physical presence, too (800 PIM to 525). Mr Neely was also a more consistent scoring threat, as evidenced by his 986 shots on goal, compared to Mr Lucic’s total at the time of his contract extension – 579. In his PY, Mr Neely was 26th in the NHL in SOG. In Lucic’s PY, he was ranked outside of the top 150. Shots on goal are a great indicator of how involved a player is in the offense, especially scoring wingers.

All else equal, more SOG means more goals and scoring chances. it isn’t fair to compare Mr Neely to more skilled and less physical players on a strictly numbers basis, it is quite apparent to see the value that Boston placed on Lucic’s intimidation and toughness.   And Mr Neely brings those same elements to the table while being a superior offensive player.

As mentioned in the team’s argument (more on that soon), it isn’t fair to present day players to compare their production to the production of players from the late 1980’s. If today’s goals-per-game number was applied to Neely’s PY production, his numbers would be very similar to Lucic’s. In a typical NHL salary arbitration case, you wouldn’t use production from 20 years ago for a comparable – again, this was just a mock arbitration hearing to give arbitrators an idea of how the process works in a hockey/sports arena.

More on the Lucic-Neely comparison – if you factor in the eras they played in, their offensive production is pretty comparable. Neely had a vastly better postseason and was a more physical presence (more PIM).

Neely Lucic

I also used SOG to indicate that Neely was improving offensively, even if his goal/assist numbers were a bit stagnant in the late 80’s (again, we had to use the assumption that we didn’t know he would break out as a 50+ goal scorer soon after).

Neely placed in the top 26 in league-wide SOG during his PY, while Lucic was ranked 149 in his PY in SOG. This was something I focused on when comparing the two players.

Neely SOG

The Team’s Case

The team’s case centred around Neely’s stagnant offensive production, the fact that he had (so far) failed to live up to his hype as a young prospect and high 1st round draft pick, and the fact that he was giving up no UFA years (Neely would have been eligible to be a UFA the summer after this one) in asking for a one-year deal.

And some context, of course, was important here:


The team raised the very important factor with multi-year contracts (like Lucic’s) -teams pay a premium to “buy” UFA years. With Neely’s arbitration hearing (a one-year contract) – no UFA years would be bought. It was a very valid and very important point to raise.


The team also introduced the thought that multi-year contracts should be evaluated on a per-year salary basis compared to the annual cap hit (or AAV). I disagreed with this, as there are countless examples of contracts with very unique year-to-year salary structures. Players are valued based on their cap hits, and that should be the number used when finding comparable salary figures.

The Team’s Comparables


The team used David Krejci and James Neal as comparables (as yielded by the above formula). Statistically, Neal was the closest comparable used by either side. However, Neal and Neely are not similar players (at least that is how I constructed my rebuttal). Big and strong, sure, and they both had/have good wrist shots, but Neely was a much more physical and intimidating player. But how is that quantified?


The Process of the Debate

Myself and my co-counsel introduced our brief, player comparables, and salary request first. Then the league had a chance to do the same with their brief, comparables, and salary request.

We were then each given time to offer a rebuttal to the other side’s brief. I spoke primarily on Krejci and Neal (and why they weren’t perfect comparables for Neely). Here are a few excerpts from our rebuttal):

Using lack of fights to indicate a lack of physicality for Neely: 

Neely’s number of fights shouldn’t be used as a way to detract from his physicality – in fact, teams often tell their good players to avoid fighting.

Why would the Canucks want Cam to spend 5 minutes in the penalty box with the other team’s fighter? He is too valuable a player for that. Hits are a great measure of consistent physical play, but unfortunately they weren’t recorded back in 1989 (and even nowadays they are inconsistently recorded from rink to rink).

Neely still put up high PIM numbers, indicating an aggressive style of play. The fact that he only fought when necessary was a skill and something he has worked on as his career has progressed

Neely was able to produce the same offensive numbers with higher PIM totals – without access to ice time numbers, that could be an indicator of a more productive player on a per minute basis.

Using Neal as a comparable:

No mention of Neal’s rapid platform year improvement due to the fact that he played with Evgeni Malkin, who led the NHL in scoring with 109 points, that season. Neal benefitted tremendously from playing with Malkin – Neal didn’t have a linemate of Malkin’s calibre during the first few years of his career in Dallas.

Many players in the past have benefitted from playing with superstar linemnates – Brendan Morrison had Todd Bertuzzi and Markus Naslund as his wingers in Vancouver (Burke liked this reference).

Neal’s production improvement could be attributed significantly to Malkin, who he played over 80% of his shifts with at even strength. Neal and Neely are both big forwards, but their playing styles are not very similar. Neal uses his size to win puck battles and create space, but he isn’t an intimidating hitter or consistent physical presence like Neely

Mr Neal has  never had more than 87 PIM in a season, while Mr Neely had 190 in his PY. In Neal’s PY, he had 108 hits – only seventh on his team,  and 157th in the entire league. Hits weren’t recorded when Mr Neely played, but he was known for being a consistent and intimidating physical presence and likely would have ranked among the league leaders.

And while Neal did sign a long term extension, it has been common for good players on top NHL clubs like Pittsburgh to take contracts at less than market value in order to build a Stanley Cup contender, as we previously mentioned with Mr. Lucic.

Comparing the plus/minus of players from different teams (and eras):

Unfortunately, plus/minus is the best statistic available to be used in arbitration cases to evaluate a player’s two-way ability. Burke brought up an interesting point after I attacked the opposition’s brief that pointed to Krejci’s superior plus/minus to indicate he was a better two-way forward than Neely (which he is/was, but it isn’t fair to compare players on different teams).

Burke wasn’t a fan of using plus/minus either, primarily because it is a stat that can be manipulated so easily to complete or construct any argument. 

And NHL teams who use proprietary statistics to evaluate players would be loathe to reveal them during the arbitration process, as they would have to turn them over to the NHLPA (and in turn the general public). So plus/minus is very limited in its value – it has use if you are comparing a player to a teammate or a linemate, but when comparing players on different teams it falls far short (even with a lot of context).

I think Mr. Burke liked my reference to his arbitration hearing with Brendan Morrison (the only arbitration case he has been involved in as a GM).

The result

We requested $7 million. The team offered $4.5. The final award was right in the middle – a one-year award for $5.75 million. Going up against Brian Burke, I considered breaking even as a huge win – even in a mock setting.

Here are some other interesting things that came up during the Q&A:

Arbitration doesn’t work in a salary capped league

The point of arbitration is to maximize a player’s worth – in a cap era, that is incongruent with building a winning team. As Burke said, asking for one more dollar means that you are taking one dollar away from  a potential teammate.

With the lower UFA age, and the death of the second contract, there simply aren’t many arbitration cases anymore.

Arbitration is a dying process in hockey

Thanks to the death of the second contract, combined with the UFA age dropping from 31 to 27, arbitration cases in the NHL are drying up. In 2004, 67 players filed for arbitration. Last summer, only 16 filed. Shea Weber was the only arbitration case that was heard during the summer of 2011.

Burke said that once a player opts for arbitration, he won’t negotiate further. A lot of agents use arbitration as a deadline to expedite the negotiating process, though.


For a significant portion of the proceedings, I was essentially debating with Brian Burke about the merits of valuing hockey players. I learned a lot about the salary arbitration process, especially during the Q&A period that followed. It was vintage Burke – he offered some great jokes and one-liners, but a lot of very interesting and relevant facts on the process of salary arbitration in hockey.