Does Doing Good Make You Play Better?
According to NHL.com, the NHL Foundation Player Award is “awarded annually to the NHL player ‘who applies the core values of hockey – commitment, perseverance, and teamwork – to enrich the lives of people in his community.’” Similarly, the King Clancy Memorial Trophy is annually “given to the player who best exemplifies leadership qualities on and off the ice and has made a noteworthy contribution to his community.” Oftentimes, it is easy to forget that hockey is, at the end of the day, merely a game, and these two awards serve to recognize players who have used their status as professional athletes to improve their communities. What struck me most about these two awards is that they are awarded entirely independently of in-game statistics. Instead, they are solely judgments of an athlete’s contributions to his community. Despite this fact, I was curious to see whether or not a player’s contribution to his community was at all linked with his professional performance.
The purpose of this article was to investigate whether a player’s in-game performance improved during the season in which he won either the NHLFPA or the Clancy Trophy. In order to do this, I looked at three rudimentary statistics, points per game (PPG), save percentage (SV%), and point shares (PS) (defined in the “Methods” section), across three seasons: the season before he won (a.k.a “Pre”), the season in which he won (a.k.a “During”), and the season after he won (a.k.a “Post”). In other words, if a player won one of these awards in 2014, I compiled his points per game, save percentage, and point shares from 2013 (Pre), 2014 (During), and 2015 (Post). Since points per game is a statistic exclusively for forwards and defensemen and save percentage is a statistic exclusively for goaltenders, each player was evaluated by examining two statistics.
From this, I decided to create two sets of comparisons. The first compared the “During” and “Pre” seasons in order to determine whether a player’s performance improved during a year in which he was contributing significantly to his community. Then, if improvement was seen, I compared the “Post” and “During” seasons to determine if this improvement was long-term or if it diminished over time.
All of the data for this project came from www.hockey-reference.com, unless otherwise cited. I ultimately decided to use the aforementioned statistics as measures for player performance as they best represent a player’s individual contributions to a game given the limitations of the website’s available data. Currently, more advanced analytical statistics exist like “Corsi For%” and “Fenwick For%” that provide a more accurate look at a player’s on-ice contribution. However, the NHLFPA and Clancy awards precede the introduction of these statistics, and as a result it would not be possible to compare all recipients.
1. On goals scored by the Oilers: approximately 30% of plus marks were awarded to Oilers who did not contribute significantly to the goal being scored.
2. On goals scored against the Oilers: nearly 50% of minus marks were awarded to Oilers who “either had no impact on the play or were doing their job defensively but nonetheless are assigned a minus mark due to an erring teammate.”
The limitations of plus-minus as an accurate indication of a player’s on-ice performance are best exemplified when looking at the NHL’s all time single-season plus-minus leaderboard. Here are the top ten members of the list, with Hall of Famers being marked with an asterisk (*):
1. Bobby Orr* (1970-71): +124
2. Larry Robinson* (1976-77): +120
3. Wayne Gretzky* (1984-85): +98
4. Dallas Smith (1970-71): +94
5. Guy Lafleur* (1976-77): +89
6. Steve Shutt* (1976-77): +88
7. Bobby Orr* (1971-72): +86
8. Mark Howe* (1985-86): +85
9. Bobby Orr* (1973-74) +84
T-10. Bobby Clarke* (1975-76) +83
T-10. Brad McCrimmon (1985-86) +83
Aside from the collection of hockey royalty on this list, most glaring is the presence of defenders Dallas Smith and Brad McCrimmon; both were respectable defensemen during their careers, but when compiling a list of all-time greats, they surely do not make the cut. How is it, then, that these two players achieved such high plus-minus totals in a single season? The answer lies in their defensive linemates from these exceptional seasons. Smith was partnered with Bobby Orr in the 1970-71 season, arguably the greatest defenseman of all time, while McCrimmon was paired with Mark Howe in 1985-86, a Hall of Famer and three-time Norris Trophy runner-up. Thus, it would not be outrageous to suggest that Smith and McCrimmon’s plus-minuses were inflated by their linemates’ performances, both of whom are also featured on this top-10 list from the same season (Orr’s single-season plus-minus is the league’s best ever, while Howe’s is good for eighth-best).
Ultimately, plus-minus is a far more representative statistic when examined over the course of a player’s entire career as opposed to a single season, which is why I disqualified it from use for this article. Interestingly enough, if we use a similar litmus test for career plus-minus, the top 20 features 18 Hall of Famers, including the likes of Orr, Gretzky, Robinson, and more. The two exceptions: Brad McCrimmon (11th best) and Dallas Smith (18th best). But I digress.
This begs the question: why use save percentage, points per game, and point shares? Save percentage (the percentage of shots faced by a goaltender that were saved) is almost entirely dependent on the goaltender’s performance, whereas statistics like goals against average, wins, and shutouts can be influenced by other factors that tie back to a team’s defensive prowess (i.e: total shots faced, penalties surrendered, etc). Points per game (the sum of goals and assists divided by total games played) was used because it allowed me to evaluate offensive performance across a season while controlling for different amounts of games played. Finally, point shares (the number of points a player contributes to his team’s end-of-season point total) were used as an indication of how said player contributed to his team winning. An important distinction must be made here: points per game refer to points awarded to an individual player when tallying a goal or assist; point shares refer to points awarded to a team after winning a game that contribute to their place in the league standings.
GP = games played
PPG = points per game
PS = point shares
SV% = save percentage
Pre award = season immediately before winning the NHLFPA or Clancy Trophy
Award = season in which player won the NHLFPA or Clancy Trophy
Post award = season immediately after winning the NHLFPA or Clancy Trophy
(Award – Pre) = the difference between statistics accumulated during the award-winning season and the preceding season
(Post – Award) = the difference between statistics accumulated during the succeeding season and the award-winning season.
Notes:
-Dough Weight (2010-11) and Trevor Linden (2007-08) both received their respective awards in their final seasons in the NHL. Therefore, their post-award statistics were not factored into the final averages.
-In 2007-08, two players received the NHLFPA: Trevor Linden and Vincent Lecavalier.
-There are five players to have received both the NHLFPA and the Clancy Trophy in the same season: Kelly Chase (1997-98), Rob Ray (1998-99), Ron Francis (2001-02), Jarome Iginla (2003-04), and Vincent Lecavalier (2007-08). Despite winning two awards, each player’s statistics were only counted once towards the averages.
-A goalie has won the NHLFPA or Clancy Trophy five times in NHL history. However, this sample size is too small to incite any meaningful analysis. I’ve included the data purely for the sake of thoroughness.
-There is no data for the 2004-05 season, as this season was cancelled due to a lockout.
Going into this article, I was hoping that the data would yield something poignant and concrete about how community service made you a better hockey player. If this were the case, then the (Award – Pre) columns would have very high, positive averages, indicating that players saw great statistical improvements in seasons where they performed a significant amount of community service. Conversely, if community service made you a worse hockey player, then the (Award – Pre) columns would have very high, negative averages. Ultimately, neither of these proved to be true.
When examining the data superficially, there appeared to be no significant improvement in a player’s performance from their preceding to award winning seasons. On average, skaters increased their point per game output over this span by a meager 0.019 PPG. When compared to the league average of 0.32 PPG in the 2015-16 regular season, that mark results in a 6% uptick, a modest improvement. The average skater played approximately 72 games per season when they won either award, meaning a 0.019 PPG increase translates to adding 1.3 points to their end-of-season total. On average, skaters only increased their regular-season point shares by 0.5 points in their award winning season. Long story short: players saw little to no statistical improvement, on average, during seasons in which they won the NHLFPA or Clancy Trophy.
However, this does not tell the whole story, as improvements can be seen from two perspectives. The first would be to determine whether a player improved statistically from a previous to award winning season, while the second would be to look at whether a player’s statistics decreased immediately after winning the award. From this perspective, a high, negative average in the (Post – Award) columns would indicate that, on average, a player’s performance dropped markedly in the season after winning their award.
This second perspective, however, operates under the (not-so-safe) assumption that players perform less community service after winning either of the aforementioned awards. While this may or may not be true (the degree to which someone performs community service is subjective, and obviously there is no data on the topic), what is clear is that only one player has ever won the NHLFPA or Clancy in consecutive years (Patrice Bergeron – more on him in a bit), indicating that the majority of skaters have not sustained an “award-winning” level of community service for multiple years. And while I understand that there are several flaws with this logic (most notably, that award recognition does not necessarily translates to the significance of a service effort), I am by no means suggesting that it is true; I’m simply examining the data and trying to see if any interesting trends have emerged (in my opinion, one has).
What the data shows is that, on average, players have markedly worse statistical years immediately after winning one of these awards. On average, players drop their points per game total by 0.13 PPG across seasons, a 41% drop. Over the span of an average post-award winning season (approximately 68 games), this translates to nearly 9 points; over a full 82-game season, it’s almost 11 points. Additionally, players saw their point shares drop by 1.7 PS after winning these awards. That point or two could be the difference between a spot in the playoffs and a spot in the draft lottery; just ask the Boston Bruins.
Speaking of the Bruins, they currently employ the only NHL player to ever win the NHLFPA and/or Clancy trophy in consecutive years: Patrice Bergeron, who won the Clancy in 2012-13 and the NHLFPA in 2013-14. Bergeron was recognized for his tremendous efforts through the Patrice’s Pals program; according to the Bruins’ website, this program “brings patients from local hospitals and other children’s organizations to the TD Garden to experience a Bruins home game…as VIP guests.” Furthermore, Bergeron’s statistical performances in these four seasons (pre-winning, two consecutive award winning, and post-winning season) are as follows: in the pre and award winning seasons, his points per game total fluctuates slightly, decreasing and increasing by 0.03 and 0.01 PPG respectively, with a 0.1 PPG drop after his final-award winning season (an 8 point drop over a full season). His changes in point shares are a bit more of an outlier; he saw a sharp decrease in point shares during his first award-winning season. However, in his second award-winning season he more-than doubled his PS total, going from 4.3 to 9.2 PS, before dropping back down to 6.4 PS in his post-award season. Ultimately, Bergeron’s four-year span is representative of the whole data set – insignificant effect in your award winning season, but noticeable change in the post-award winning season.
So back to the titular question: does doing good make you play well? To put a long story short: I’m not really sure. Nothing indicates that a player will perform better in a season where they exhibit outstanding community service, although there is some data to suggest a correlation between performing poorly in the season following winning a community service-based award. However, this should not discourage community service; it should do quite the opposite, in fact. Players should see this as a motivation to sustain any and all community service efforts to fend off statistical detriments. What is also encouraging is that there is no data to suggest that community service decreases a player’s statistical performance, only that the potential lack of community service would do so.
However, it is important to remember that community service should not be done for any statistical benefits, but rather to try and improve the lives of others less fortunate. While hockey players may not be the most financially successful athletes compared to their athletic counterparts in Major League Baseball or the National Basketball Association (an issue for another day), the NHL does have a very active and passionate fan base. Professional hockey players are kept in a special place in the hearts of many fans across the country and throughout the world, and with this adoration, these athletes have a unique ability to touch their fans in a way few others can. While I doubt this article will receive a following strong enough to reach the eyes and ears of any professional athletes, I hope that these players are able to come to this realization on their own and continue to serve their communities admirably.