|
Who were your favorite players growing up? Probably, the players whose careers you followed most devotedly from team-to-team, whose statistics you can still recite and whose hockey trading cards you still have tucked away in a shoebox somewhere.
One of mine was Mike Gartner. He was from my old hometown of Ottawa, Ontario, and he was absolutely amazing. He flew down the ice like the wind and was almost impossible to keep away from the net. He bagged 30 goals a season like clockwork, his record of 15 consecutive seasons of 30 or more goals still stands today. Sadly, he never got to play for the Stanley Cup and is probably the greatest player never to win an individual award.
I would love to see Mike Gartner play again, just as you would probably love to see your favorite players in action again. Unfortunately, your favorite players are probably retired. However, we could use reverse career projections as the next best thing.
In order to find the next Mike Gartner, let's take a good look at his career NHL statistics, including running totals. This is all the information we'll need to scour today's NHL to find some modern-day equivalents.
Season GP G A PTS PIM GP G A PTS PIM
1979-80 77 36 32 68 66
1980-81 80 48 46 94 100 77 36 32 68 66
1981-82 80 35 45 80 121 157 84 78 162 166
1982-83 73 38 38 76 54 237 119 123 242 287
1983-84 80 40 45 85 90 310 157 161 318 341
1984-85 80 50 52 102 71 390 197 206 403 431
1985-86 74 35 40 75 63 470 247 258 505 502
1986-87 78 41 32 73 61 544 282 298 580 565
1987-88 80 48 33 81 73 622 323 330 653 626
1988-89 69 33 36 69 73 702 371 363 734 699
1989-90 79 45 41 86 38 771 404 399 803 772
1990-91 79 49 20 69 53 850 449 440 889 810
1991-92 76 40 41 81 55 929 498 460 958 863
1992-93 84 45 23 68 59 1005 538 501 1039 918
1993-94 81 34 30 64 62 1089 583 524 1107 977
1994-95 38 12 8 20 6 1170 617 554 1171 1039
1995-96 82 35 19 54 52 1208 629 562 1191 1045
1996-97 82 32 31 63 38 1290 664 581 1245 1097
1997-98 60 12 15 27 24 1372 696 612 1308 1135
We can use Similarity Scores to find today's Mike Gartner's from these basic statistics. Here's how we proceed:
For each player in the NHL last season, we calculate a Similarity Score against each of Gartner's 18 seasons beyond his first. This Similarity Score compares the current season, the previous season, and the career totals prior between Gartner and each active player's most recent season.
Having calculated a Similarity Scores for each active player against each of Gartner's season, we take the minimum of the 18 Similarity Scores to match each individual player against the point in Gartner's career that's most applicable to them.
The top matches will be whoever has the lowest Similarity Score against any of Gartner's seasons. In essence, they will be the players that most closely match Gartner at some point in his career. Let's look at the top two matches.
Dany Heatley GP G A PTS PIM
Prior 272 130 154 284 218
2006-07 82 50 55 105 74
2007-08 71 41 41 82 76
2008-09* 62 30 28 58 72
Ilya Kovalchuk GP G A PTS PIM
Prior 305 160 143 303 216
2006-07 82 42 34 76 66
2007-08 79 52 35 87 52
2008-09* 64 35 37 72 35
It's not surprising to find Dany Heatley and Ilya Kovalchuk at the top of the list. The former linemates are both extremely fast and exciting players to watch and share Gartner's same knack for scoring. I find it quite uncanny that they're already among my favorite players today. The only significant difference between them and Mike Gartner appears to be that they've both already won individual awards, and one of them has already competed for the Cup.
Speaking of which, Mats Sunin showed up on the list. There's another player I love, and who is known for his Gartner-like consistency, scoring 70 points or more in every single full season after his first. Sundin shares the distinction of also never having competed for the Stanley Cup, and his amazing career has somehow managed to be kept almost free of individual award (twice a 2nd-team all-star). If there were a “Mike Gartner” award, Sundin would undoubtedly have a closet full of them.
Mats Sundin GP G A PTS PIM
Prior 1231 523 720 1243 989
2006-07 75 27 49 76 62
2007-08 74 32 46 78 76
2008-09* 22 6 9 15 18
I always wondered how well Mike Gartner would have played had he continued for just another season or two. The two players today that most closely matched Gartner near the end of his career are Mike Modano and Jeremy Roenick. By watching them we can get a taste of how Gartner would have produced had he played through until the end of the century.
Mike Modano GP G A PTS PIM
Prior 1083 465 632 1097 730
2006-07 59 22 21 43 34
2007-08 82 21 36 57 48
2008-09* 62 15 25 40 36
Jeremy Roenick GP G A PTS PIM
Prior 1252 495 675 1170 1413
2006-07 70 11 17 28 32
2007-08 69 14 19 33 26
2008-09* 33 3 7 10 22
Evgeni Malkin, Marian Gaborik, Olli Jokinen, Teemu Selanne and Eric Staal also showed up near the top of the list.
There may be only one Mike Gartner, but I'm quite satisfied with the list and I look forward to future explorations to find the next Brian Trottier or Ray Bourque. It will be interesting to see what we come up with as player comparables for those two.
Today we've demonstrated that Similarity Scores can be useful for more than just establishing career projections for active players. For example,
What if your favourite player was Pat LaFontaine, Al Iafrate, Cam Neely or anyone else whose career was cut short by injury? Find similar players, and study how their careers progressed from there.
If you think your favorite player is underpaid, we can find similar players and compare salaries.
We can now also find the next Gilbert Perreault.
While I head off to eBay to bid on some Dany Heatley and Ilya Kovalchuk merchandise, I'll leave you with one final question to ponder: What other uses can you think of for Similarity Scores?
Afternote: Hockey historians probably noted that I ignored Mike Gartner's single season with the Cincinnati Stingers of the WHA. He scored 27 goals and 25 assists in 78 games playing with Mark Messier.
Season GP G A PTS PIM
1978-79 78 27 25 52 123
How significantly can we improve our career projections if we included the WHA [ed. note - yay WHA!] in our historical database? Could it help us find the next Andre Lacroix, Marc Tardif or Real Cloutier? Stay tuned!
Robert Vollman is an author of Hockey Prospectus.
You can contact Robert by clicking here or click here to see Robert's other articles.
|