Keen students of the Copper & Blue masthead will recall that yesterday morning, Scott Reynolds and Benjamin Massey faced off in a faux arbitration hearing to determine the future contract value of one Gilbert Brule, an impending restricted free agent coming off the best season of his disappointing career. It was difficult dragging the two of them out of their other projects to work on this task, as Scott was running complicated, fascinating analyses on scoring chances in the Stanley Cup playoffs and adding immeasurably to the depth of knowledge we have about playoff hockey, while Ben was working on equally important statistical problems like "if Mike Comrie, Liam Reddox, and Sam Gagner were on a line, would they have the highest cumulative jersey numbers in NHL history?"
However, having been thrust into the argumentative fray, Scott and Ben didn't let up with just one presentation disputing Brule's rightful level of compensation. On the contrary, they spammed the Copper & Blue mailing list with debate, statistics, fascinating insight, and unprintable comments about each others' mothers. At last, cooler heads prevailed and the two warring authors were persuaded to set their differences aside long enough to bring their rebuttals out to a public forum where the Internet can
mock marvel at them.
What follows is the second part of Gilbert Brule's mock arbitration hearing. The series will conclude tomorrow when Bruce McCurdy will weigh the facts and come to an honest, impartial decision. As always, feel free to make Bruce's life a little easier (and these guys' a little harder) by chiming in with a comment on whatever elementary error of logic or mathematics we've committed this time.
A Rebuttal in Defense of Gilbert Brule (Benjamin Massey)
My opponent, Mr. Scott Reynolds, raises a number of interesting points in his bid to have a seventeen-goal scorer paid less money than Donald Brashear, Georges Laraque, Wayne Primeau, or friend of yours and mine Matt Cooke. Reynolds lumps Brule in with the cementheads and the entry-level contracts without heed to the fact that Brule, in fact, has an NHL skill player's resume and a too-rare ability to get results even when dealing with a truly God-awful hockey team in the Edmonton Oilers.
The mistakes Reynolds makes are quite thorough. Much of his argument rests on goal differential numbers: Brule's -0.37 EVGD/60 against Ryan Callahan's 0.62, for example. But goal differential is affected by line mates, goaltending, quality of opposition... it cannot be used as a simulacrum of an individual player's performance relative to others in the NHL. His team as a whole finished with a -70 goal differential last season, which is the main reason they were dead last in the National Hockey League. Dustin Penner and the remarkable Zack Stortini were the only Oiler forwards to play over thirty games and record a positive EVGD/60. Brule's figure was markedly superior to that of Mike Comrie, Ethan Moreau, Shawn Horcoff, Patrick O'Sullivan, Robert Nilsson, and Fernando Pisani among those who have been in the NHL long enough to have burned off their entry-level contracts. Even the golden boy, Sam Gagner, heralded as a saviour of the franchise, had a -0.41 EVGD/60. Bearing in mind the context, Brule's pedestrian -0.37 is in fact a strong if not sterling number in spite of the cherry-picked examples Reynolds attempted to discredit him with. Reynolds makes a similar mistake comparing Brule's QUALCOMP and QUALTEAMs: trying to compare such figures across teams, never mind across conferences as Reynolds does, is simply line noise.
Nobody disputes that, as Reynolds accurately shows, Gilbert Brule's career numbers are bad. But Brule is the poster child for a player who was mismanaged in his youth. His travails in the Columbus Blue Jackets oragnization have been well-documented, and unlike Rick Nash he didn't wind up with Ken Hitchcock to save him. Brule is not the first player whose value the perennial also-ran Blue Jackets have distorted. Nothing in Brule's percentages or rate numbers are superhuman and suggest he can't sustain this pace. On the contrary, since joining the Oilers Brule was one of the few bright spots on the 2008-09 Springfield Falcons and had a promising eleven-game call-up where he showed more-or-less this level of ability.
Finally, there is the discussion of Brule's problems on special teams. There is no use in sugar-coating it: in the small sample sizes we saw of Brule on the power play and the penalty kill indicate a player who is not a contributing NHL-calibre player 5v4 or 4v5. But even strength play is the heart of the game and deserves to be valued as such. Rare is the game that's not played at five-on-five two-thirds of the time, which is why statisticians place such emphasis on getting a player's even-strength numbers when analyizing the player. Brule is not a major power play contributor and his statistics last season reflect that. His contract in 2010-11 should of course reflect that as well. But Brule's ability to contribute at even strength is vastly more important than any degree of power play proficiency could ever be. Having played only 14:14 a night last season, Brule has plenty of rooom to grow into an expanded role and increase his value to the Oilers, or whichever NHL team recognizes his value, even further.
Rebuttal Presented By The Edmonton Oilers Hockey Club (Scott Reynolds):
The Edmonton Oilers disagree with several comments that have been made by Mr. Brule's representative. The first issue we would like to address is the discussion around Gilbert Brule's percentages and the sustainability of his performance. Mr. Brule's representative stated that "bang[ed] in seventeen goals on an awful hockey club without even posting a conspicuously obscene shooting percentage." Although the first half of that statement is perfectly true, the last half is not. I am happy to agree with Mr. Brule's only significant contribution to the hockey club came at even strength. Looking at those numbers closely, we see that Gilbert Brule's shooting percentage at even strength was 17.0% in 2008-09, a number so much higher than what he has personally accomplished in the past and so much higher than the league average of 8.7% that it is almost certain to regress downward as was mentioned in our original arguments. Mr. Brule's representative briefly mentions our concerns about the sustainability of this performance but does not offer any evidence to suggest that the performance will in fact be sustainable. He does not mention Brule's higher than average on-ice shooting percentage or his above-average individual point percentage. His claim that there is "no reason to believe he can't sustain [his offensive results]" is false.
The next portion of the argument set forth by Mr. Brule's representative that we take issue with are the claims around his ability to draw penalties. Mr. Brule's representative stated that "[Gilbert Brule] draws more penalties than usual for a player at his position while taking just thirty-eight minutes in penalties himself." The first half of this staement is false and the second half is misleading. In fact, Gilbert Brule draws fewer penalties (he draws 0.9/60 minutes) than the average forward (1.1/60) and his penalty differential is also inferior (0.1/60) to that of the average forward (0.22/60). The claim that Gilbert Brule draws more penalties than average is false. In fact, Gilbert Brule's inability to draw penalties is a reason to decrease his salary relative to his peer group, not increase it.
We would also take issue with Mr. Brule's representative in his claims about outscoring. He stated that "[Gilbert Brule] did not outscore but he came as close to doing it as any regular forward not named Dustin Penner." I suppose it depends on one's defiinition of "regular" but of the forwards who appeared in at least twenty games with the Oilers this season, Gilbert Brule placed behind Ales Hemsky, Dustin Penner, Ryan Stone, Zack Stortini and Andrew Cogliano in terms of +/- per sixty minutes of ice time. In that Cogliano and Stortini each played more than 75 games, I think it is fair to say that this statement from Mr. Brule's representative is false.
Mr. Brule's representative also mentions several potential comparables to Gilbert Brule. The only comparable he mentions with a salary higher than his suggested award of $2.7 million dollars is Mikhail Grabovski. However, Grabovski is not a good comparable for Mr. Brule for several reasons. The first is the length of his contract (three years) combined with the age of the player when the contract was signed (he had just finished his 25 y/o season). These are both significant factors in the amount given to Grabovski. Further, Grabovski is a superior offensive player. His point production in his platform year is superior to that of Brule (0.62 pts/game compared to 0.57 for Brule) and his offensive performance at the AHL level in his career is also far superior to that of Brule (0.95 pts/game in 78 games compared to 0.62 pts/game in 55 games). Grabovski's career performance at the NHL level prior to signing his contract was also far superior to that of Brule (0.54 pts/game compared to 0.32 for Brule). Moreover, Grabovski played a significant role on the Toronto power play (2:06 per game) which makes his role signficantly different from that of Brule. Given the questions around the sustainability of Brule's performance, the Oilers do not feel that Grabovski - a superior offensive player - is a good comparble, especially given the differences in circumstances (length of contract, age, role on the team) between the two players.
Mr. Brule's representative also compares Mr. Brule to Ryan Callahan, as we also have done. In our brief we suggested that although the two players were comparable offensive players in their platform years, that Callahan is the superior player. Mr. Brule's representative describes Callahan as a "pudding-soft players who [is a] complete non-entit[y] in [his] own zone." Yet Callahan faced more difficult opposition than did Brule. He was also trusted with a higher percentage of his team's defensive zone faceoffs than was Brule and had a superior goal differential. He is also entrusted with a regular role killing penalties. Mr. Brule's representative gives no actual evidence to back up his claim that Ryan Callahan "[is a] complete non-entit[y] in [his] own zone." Offensively, I will remind you that Ryan Callahan has a better long-term track record than Gilbert Brule. His career points per game is superior and his AHL points per game is superior. Furthermore, Ryan Callahan has a much better penalty differential. If either player is going to force the opposition "into doing something it'll regret" it seems that it's far more likely to be Callahan. Mr. Brule's representative concludes by saying that "nobody can compare Ryan Callahan and Gilbert Brule and say that even if they get the same number of points (and Brule will get more) they'd be equally valuable players," the implication being that Gilbert Brule is the superior player. I hope that the above paragraph demonstrates that, if anything, the opposite is the case, that it is Callahan who is unquestionably the better player. Finally, there is no evidence presented for the claim that Brule will get more points and given the offensive histories of both players, it would seem Callahan is the better bet.
The last comparable used by Mr. Brule's representative is Robert Nilsson. Nilsson would actually have made the list of comparables that we used in our original submission if not for the length of his contract. In that the contract was signed at a young age and is only one year longer than the two-year limit that was set, we agree that this player is a reasoanble comparable. As such, I will include the complete chart comparing the two players:
In terms of the contract, we see that Nilsson's $2 million dollar contract fits within the already established range of $775,000 and $2.3 million whereas Mr. Brule's representative has proposed an award well outside of this range. It continues to be our belief that Mr. Brule belongs somewhere toward the bottom end of this range. As we have seen with the other comparables, once again Robert Nilsson is the superior player:
Robert Nilsson is clearly the superior player here. His goal differential is more than a full goal per sixty minutes better than that of Gilbert Brule. His penalty differential is, much like the other comparables, significantly better than that of Brule. He started more often in the defensive zone. He achieved his results against lesser competition than Brule but also did it with inferior teammates. In fact, for much of the second half of the season Robert Nilsson was the veteran leader on a line with two raw rookies (i.e. no professional experience of any kind).
Once again, Gilbert Brule shows himself to be the clearly inferior offensive player at the career level and still leave him in sixth of six in terms of outscoring. This data taken together demonstrates that, although the even strength scoring is similar in their platform year, Nilsson is the clearly superior player.
In conclusion, we have added one player to the list of comparables who fits within the already established salary range for this group but we do not feel that this should alter Gilbert Brule's award. Our original proposal of $1.125 million dollars fits within this established range, whereas the $2.7 million dollar award does not and is nothing but blue-skying. The $1.125 million dollar award gives Brule a raise of 41% which is commensurate with the raise given to the player at the low end of the group, Tyler Kennedy. It also puts Brule's salary on par with another member of the comparison group, Daniel Paille. Given that Brule is the weakest member of the comparison group in most categories it does not make sense to give him a salary that would put him at the high end or even in the middle of the comparison group. We think that the proposed $1.125 million dollar per season offer is very generous.