I’ve been on Lowetide’s show and a podcast, Toronto Sports Media Pressbox a bit lately and referenced how the salary structure has changed in the NHL since the lockout without really having yet backed it up in great detail. I thought I’d take a little time and do that.
I’ve taken all of the forwards from 2003-04 and 2011-12 who played at least 40 games in that season and sorted them by TOI/G. I’ve then broken them into groups for comparison purposes – top 90 are “first liners”, second 90 are “second liners” etc. I’ve then dropped in their salary information, which permits me to prepare some graphs to illustrate which groups of players have made out well since the lockout and which haven’t done so well.
A key piece of information that’s useful for contextual purposes. Since 2003-04, NHL revenues have increased at about 5.8% per season, even counting 2004-05, going from $2.1B to $3.3B. Let’s get into the graphs.
Unsurprisingly, no group of forwards has experienced a collective annual increase in wages that matches the rate at which revenues grown. This is a reasonably sophisticated point and one that Greg Wyshynki’s “The Player” indicated that Don Fehr had been making to players as they prepared for this round of collective bargaining talks.
In 2003-2004, the final season before the lockout, players’ salaries ate up about 74 percent of all the revenue generated by the NHL. In that year, the business was worth about $2 billion.
As we know, in 2011-2012 players received 57 percent of the revenues, which are believed to be in the neighborhood of $3.3 billion. So, over the course of that time revenues have increased by over 50 percent while player costs have only increased by about 15 percent.
Note as well that the increase in the median salary in the four groups that I’ve defined has been much more significant than the increase in the average salary. Now look at this:
You can see that salaries have increased but that the standard deviation has gotten smaller. What’s happening, as you’ll see in the coming graphs, is that the various groups of players are becoming more tightly grouped in terms of how much money they earn. The spread in terms of first line incomes is smaller than it was in 2003-04 and so on.
I wanted to dig a bit deeper into this, so I sliced my data some more. I took each group of players (first liners, second liners etc.) and sorted it by the salaries that those players received. I then looked at how that’s changed from 2003-04 to 2011-12. We’ll start with the first liners.
The graph that you’re looking at graphs three separate things: how much the ten players in each group earned on average in 2003-04 and 2011-12 and how much they would have earned with 5.8% increases from the amount that they were earning in 2003-04. So, to illustrate, in 2003-04, the ten highest paid forwards with first line ice time were Jaromir Jagr ($11MM), Sergei Fedorov ($10MM), Keith Tkachuk ($10MM), Joe Sakic ($9.88MM), Mike Modano ($9MM), Mats Sundin ($9MM), Bill Guerin ($8.86MM), Bobby Holik ($8.85MM), Doug Weight ($8.5MM) and Jarome Iginla ($7.5MM). They earned an average of $9.26MM.
Last year, the ten highest paid first line forwards in the NHL with at least 40GP were Brad Richards ($12MM), Vinny Lecavalier ($10MM), Evgeni Malkin ($9MM), Alex Ovechkin ($9MM), Steven Stamkos ($8MM), Dany Heatley ($8MM), Joe Thornton ($8MM), Jason Spezza ($8MM), Marian Hossa ($7.9MM) and Eric Staal ($7.75MM). That’s an average of $8.77MM. If the salaries of those ten players had grown at the same rate as NHL revenues since 2003-04, they would have earned an average salary north of $14MM.
In fact, none of the thirty highest paid first liners have seen their salary increases keep pace with the NHL’s 5.8% growth rate. The 31st to 81st highest paid players however, all have had their salaries match that 5.8% growth rate. The 81st to 90th highest paid first liners haven’t, but then that group tends to consist of guys on entry level contracts. The takeaway? The first liners who have benefitted from this CBA tend to be guys who would have been towards the lower end of the first line pay scale pre-2005.
Let’s talk about Taylor Hall for a second. As you’ll be aware, he’s just signed a contract for $6MM per for seven years. On last year’s numbers, that would put him in the top thirty in NHL salaries. We’ve just finished a CBA that seems to have imposed considerable restraint on the growth of the salaries of the highest paid (and, presumably, best) of the first line players. There are two things that I’m reasonably comfortable in thinking Taylor Hall will do over the next eight years: make hilarious #OHLeducation mistakes on Twitter and be one of the thirty best forwards in the NHL. For that reason, if I was the Oilers, I’d be pretty comfortable with his contract. Hall was tied for fifteenth in the league in G/G last year, playing on a bad team, driving the bus at ES and not getting the shelter that some of the other young Oilers got. He’s going to be a great player. Signing him during a period in which the really elite salaries aren’t growing makes great sense to me.
Which leads to Jordan Eberle, a player I’m less comfortable with the Oilers signing long term. Let’s look at the second liner graphs before getting into that.
In some ways, this is very similar to the first liners. The top twenty highest paid second liners haven’t seen growth matching that enjoyed by the league as a whole. As with the first liners, the highest paid second liners are actually now making less than they were in 2003-04. The 21st through 60th guys have done better than the league average growth since 2003-04.
Look at that first chart, the salary growth chart. You can see that there aren’t a lot of second liners getting Taylor Hall money, which is presumed to be in the neighborhood of what Eberle’s people are hoping is Eberle money. A lot of the guys in the first group have contracts that are probably perceived of as being albatrosses by the teams that signed them: Danny Briere ($7MM), Alex Semin ($6.7MM), Mike Cammalleri ($6.6MM), Thomas Vanek ($6.4MM), Ville Leino ($6MM), Ryan Malone ($5.5MM), Tim Connolly ($5.5MM), Johan Franzen ($5.25MM), Ales Hemsky ($5MM) and RJ Umberger ($4.5MM). It’s probably not a coincidence that Semin moved on as a UFA, Vanek was an offer sheet and Cammalleri was traded during his contract. Nobody hopes to spend that sort of money on a second line player. You can look at all of those deals and understand how the team signing them hoped to be getting a first line player.
Back to Eberle. What’s awkward about him is that there’s plenty of reason to think that he isn’t going to be able to get first line ice time in Edmonton in the future. Competition on the right side on the Oilers could be insane as early as 2013-14, with a hopefully healthy Hemsky, a sophomore Nail Yakupov and Eberle all fighting for minutes. Even if you look at the first line numbers, you can see that the big money, the $6MM type deals, is reserved for guys who teams think are going to be superstars, or who are superstars. If you pay a guy who can put up 50-60 points with first line ice time that sort of money, which is a more reasonable expectation for Eberle, the market says you’re doing it wrong.
More broadly, again note the swell in salaries for the middle paid group here. Again, salaries have risen substantially but those benefits haven’t really been directed to the higher end second liners – it’s been that middle swell that has increased salaries.
Of course, the price that the market puts on a player and the value of his actual contribution are not at all the same thing. With that said, as a fan of the Oilers and someone who wants them to do well, I am awfully concerned that the swell in the price of mid-priced (and by implication, middle of the road talent level) first and second liners is not due to the market concluding that it was previously pricing these players incorrectly but due to the need of all teams to spend within the salary range. In the pre-2005 NHL, teams could run with first and second lines made up of cheap young players and keep their payrolls down while wealthier teams spent way more money on older, more expensive players; as that’s no longer an option and teams have been forced to spend money on players, spending money on your top end players, regardless of whether they deserve it relative to other team’s top-end players seems like one semi-sensible approach.
If that is what has happened, and the new system, whatever it might be, takes away those pressures, you might reasonably expect to see stagnation or decline in the price of those players. In those circumstances, I’d rather the Oilers weren’t locked into Eberle at 2012 prices for mid-level first line players or second line players.
Third liners haven’t really enjoyed the growth in the NHL’s revenues that’s been enjoyed by the mid-level first and second line players. The only sub-group of them that’s beat the 5.8% league growth number are the lowest paid, who have just barely beat it. Again, you see the highest paid third liners actually making less now than in 2003-04. It’s interesting to me how low the salaries of the mid-level group of them actually are – the 31st to 60th highest paid third liners are averaging $1.26MM compared to $2.45MM for the same group of second liners; they average about 84% of the TOI. 84% of the TOI for 51% of the money. When you consider that teams are locked in an efficiency contest, looking for ways to spend their money as efficient as possible, if you accept that the difference between second and third liners isn’t massive, you wonder if there’s an advantage to having slightly worse second liners to purchase slightly better third liners.
The fourth liners have really been ground down. The highest paid thirty of them have seen their salaries go backwards since 2003-04. None of them have matched the growth of league revenues. If we throw them into the mix of the middle group of second and third liners that I described above, you see that they’re getting about 65% of the money paid to third liners with about 71% of the TOI; when you consider that third liners are generally getting more in terms of responsibility and competition, that doesn’t seem unreasonable.
What does it all mean? Well, as I’ve said before, there was no monolithic player loss in the last lockout. Certain groups of players have done much better than they might reasonably have expected to. In particular, this includes mid-level first and second line players. If I were to do this again by age, we’d likely see that many of those were guys who would have been RFAs under the old CBA. Their gains have been paid for by the better players on first/second/third/fourth lines, all of whom have seen their average salaries go down and by third and fourth lines more generally, who have not enjoyed salary appreciation in line with the league average.
One final point I wanted to mention – this information makes what the Phoenix Coyotes have done all the more interesting. Their guys who played first line ice time by my definition – Shane Doan, Ray Whitney, Martin Hanzal and Radim Vrbata – were all really cheap first liners, with only Doan costing them more than $3MM per year. They had one very expensive second liner – Daymond Langkow – and then Boyd Gordon and Lauri Korpikoski, both of whom came from the lower end of the second line price range, costing $1.3MM and $1.75MM respectively. Their only player to play third line ice time fell in the middle of the salary range for third liners. Then they had Raffi Torres and Taylor Pyatt, both of whom played fourth line minutes (although it was close) and acquitted themselves well but who were pricy for bottom of the lineup players.
In effect, the ‘Yotes picked their places to spend their money and seemed to focus more of their forward spending relative to their opponents at the bottom of the lineup. They should stand as an example to NHL teams as to what you can do if you devote yourself to finding ways to generate a competitive edge – there’s no reason teams with large budgets can’t ask themselves where they can get more bang for their buck rather than just throwing three years and $1.75MM per at Eric Belanger, just because the staticstial analysis says he’s good. Even if you can’t quantify it precisely, it still seems to me to be a healthy framework in which to think about these things.