Sunday, May 24, 2009

Windsor wins Memorial Cup, scores one for Malcolm Gladwell

The Windsor Spitfires, junior hockey champions, were a great third-period team.

And a great second-quarter team.

The wordsmiths covering the Memorial Cup can capture the spirit of Windsor's triumph for the papers and the sportscasts. The Spitfires, representing a city devastated by the bottoming-out of the U.S. auto industry, are the first team to win the Memorial Cup after losing its first two games of the tournament. They are just a year removed from captain Mickey Renaud's death from a rare heart ailment age 19, which GM Warren Rychel said at the time was "the biggest tragedy in Spitfire history."

(Puck Daddy has did it justice, brilliantly.)

A broader if not necessarily larger truth could be Rychel and coach Bob Boughner, who as our own Trevor Stewart explained last November, have a special eye for talent. This is kind of a complicated post and you're excused for not reading.

The book Outliers shone a light on the disproportionate number of elite hockey players born in the first three months of the year. The Spitfires actually had the fewest (7) of the four teams at the tournament. The tourney MVP, good Kingston boy Taylor Hall, is a November baby, who is not eligible for the NHL draft until 2010. Windsor's captain and No. 1 goaltender were also born late in the year. The runners-up, the Kelowna Rockets, whose Western Hockey League drafts players at age 15 (a year earlier than the other two leagues), had the most, 10, which is fairly typical. The point is the obvious: Maybe one way to get a read on a major junior team's front office is to see how many regulars it has who were born outside the Jan. 1-March 31 window.

This does not explain the outcome of today's game. That debate could be steered around to the ridiculously long layoff Kelowna had between meaningful games. The Spitfires scored on their first three shots before the Rockets could even skate off the rust from four days between games, for pity's sake. As defenceman Ben Shutron told Gare Joyce of, they were "the best-conditioned team in the tournament."

However, what made that possible? They had to get there first. Looking at the rosters for the two finalists hints at Malcolm Gladwell's point in Outliers about how we are not "always particularly smart about how to make the best use of our talent." (, Dec. 8, 2008.) It might also illustrate how Rychel/Boughner are smarter than most of their competition.

As Gladwell explains, there is a huge blind spot with identifying good hockey players. The major junior teams which can negotiate around this might be the ones who put themselves in position to pull off something like Windsor did:
"It's a beautiful example of a self-fulfilling prophecy. In Canada, the eligibility cutoff for age-class hockey programs is Jan. 1 ... Coaches start streaming the best hockey players into elite programs, where they practice more and play more games and get better coaching, as early as 8 or 9. But who tends to be the 'best' player at age 8 or 9? The oldest, of course — the kids born nearest the cut-off date, who can be as much as almost a year older than kids born at the other end of the cut-off date. When you are 8 years old, 10 or 11 extra months of maturity means a lot."
The Windsor brain trust's wisdom might be borne out when you group the Spitfires' and Rockets' Memorial Cup rosters by which quarter of the year the players were born in (January-February-March; April-May-June; and so on.)
Kelowna: 10-4-6-3
Windsor: 7-10-4-5
Kelowna's 10 players born between Jan. 1-March 31 include star defenceman Tyler Myers (Feb. 1, 1990) and starting goalie Mark Guggenberger (Jan. 10, 1989). The Rockets were considered an improbable WHL champion, but the team they upset in the championship series, the Calgary Hitmen, had a somewhat similar birthday distribution: 10-6-4-2. The main reason to cite Kelowna and Calgary is to show what the age distribution often tends to be with major junior teams: Lots of players born early in the year, very few born in the summer and fall. It doesn't explain why Windsor won, it's just background for how they rebuilt in two years.

At age 16, when players are drafted in the OHL, the advantage early-birthday kids have can wreak havoc with teams' evaluations. It evens out somewhat by the time those players are 18, 19 and 20. The Windsor team which Warren Rychel assembled has several contributors born in the second quarter of the year, between April 1 and June 30 (hence second-quarter team):
  • Greg Nemisz (June 5, 1990) — Nemisz, who had seven points in six games at the Memorial Cup, was thought of as a bit of reach when he was taken with the No. 7 overall pick in the 2006 OHL draft. Two years later, the 6-foot-4, 200-lb. winger became a first-round pick in the NHL (25th overall to Calgary).

  • Dale Mitchell (April 9, 1989) — Like Nemisz, Mitchell went in the same round of the NHL draft that he did in the OHL draft (third round to the Leafs in 2007 and third round by Oshawa in '05). Mitchell, a trade deadline pickup, scored a natural hat trick in the tie-breaker game vs. Rimouski and some thought he, not Hall, could have been the MVP.

  • Andrei Loktionov (May 30, 1990) — Also notched seven points during the tournament. Kind of a wild card since he's Russian.

  • Mark Cundari (April 23, 1990)— One of Windsor's top defencemen; was able to take a regular shift in the OHL as a 16-year-old rookie in 2006-07, when Boughner and Rychel went with a youth movement.

  • Ben Shutron (June 14, 1988) — A feel-good story, since he played for Kitchener in the '08 Memorial Cup and was a whipping boy while he was with the human comedy which is the Kingston Frontenacs. Gare Joyce wrote a nice piece about Shutron for ahead of Sunday's final.
Windsor was the only team at the tourney with more April-May-June birthdates than Jan.-Feb.-March (Drummondville, the Quebec league champ, had eight of each).

Meantime, three of Windsor's veteran role players were all "wrong birthdays" who were acquired in trades:
  • Rob Kwiet (Aug. 2, 1988) — An overage defenceman who started out with the Mississauga St. Michael's Majors. Scored the third Windsor goal.

  • Lane MacDermid — A tough guy who was a valuable penalty killer during the playoff run. Originally taken 159th overall in the 2005 draft by the Owen Sound Attack.

  • Scott Timmins (Sept. 11, 1989) — Centre who was with Kitchener last season. Originally a fifth-round pick.
Last but not least, there are the fourth-quarter players:
  • Andrew Engelage (Oct. 28, 1988) — A kind of OHL version of the Red Wings' Chris Osgood (often criticized, but his teams win) who was originally a 13th-round draft choice. Ended up being in net for 64 wins this season after Windsor was criticized in some corners for not trading for a more ballyhood goalie, such as Red Wings prospect Thomas McCollum.

  • Taylor Hall (Nov. 14, 1991) — One wonders what Hall's progress would have been if his family had not moved from Western Canada to Kingston before he was exposed to the draft. Hall's father was a pro athlete (CFL), which means he has the genes to counter his wrong birthday.

  • Harry Young (Nov. 12, 1989) — The team captain and a Windsor native, Young was a fourth-round pick in 2005 by Guelph. Windsor got him via trade.
Of course, by this point, you might have figured out, "Hey, smartass, Windsor's best NHL draft prospect, Ryan Ellis, was born Jan. 3! Doesn't that blow the theory out of the water?"

Ellis, who just became the 20th player to win a world junior gold medal and a Memorial Cup in the same season, is an outlier. Concerns about his size (5-foot-10, 173 lbs.) led to the offensive defenceman slipping through the entire first round in the 2007 OHL draft before Windsor snapped him up in the second. If you read the feature story Sun Media's Ryan Pyette wrote about Ellis ahead of today's game, you see traces of that "ten thousand hours" theory:
" 'I was constantly shooting pucks,' Windsor Spitfires star defenceman Ryan Ellis said of growing up in Freelton, Ont., near Hamilton. 'We had a (white) garage door my dad tried to repair a million times. But it didn't work. There was really nothing left of it. It had brick in the middle but it just covered the net.

" 'I basically destroyed the door.'

"... Boughner remembers, a couple of years ago, pulling into the Ellis driveway to interview the Spitfires' potential draft pick. He saw the garage door with the terrible dents.

" 'Warren (GM Warren Rychel) said I have to see this kid play and after I did, I went to meet him,' Boughner said. 'You watch him knock down bouncing pucks at the blue line and he has the most unbelievable hand-eye co-ordination I've ever seen. It's innate talent, but he works harder than anyone at it. He shoots a ton of pucks. In the games, he doesn't shoot enough for my liking. I'm always on him to shoot more.' "
You know the rest of the story. Ellis scored the sealer today, the fourth goal in a 4-1 win, on a slapshot from the point after some nice work along the end boards by the November baby, Taylor Hall.

Obviously, junior teams should not draft or trade based on birthdates. Maybe they should use it a check when they try gauge the upside of a 15- or 16-year-old player. (Oddly enough, the Belleville Bulls are losing P.K. Subban, who went from sixth-round pick in the OHL to a top prospect for the Montreal Canadiens and a two-time world junior defenceman. Subban's birthday? May 13, making him another second-quarter player.)

All this might go to show is how well each team stickhandles around a built-in bias in hockey toward players born in January, February and March. It might show how well teams do with evaluating a player who is not a finished OHL-ready product at age 16, which is almost every player.

It is interesting if you look at the birthdates of players on an also-ran team. Selecting one at random ... the Kingston Frontenacs.

It turns out the Frontenacs are absolutely top-heavy with January, February and March birthdates, 12 in total. In the 2007 OHL priority selection, they didn't take a single player born between July 1-Dec. 31. Coincidentally or not, the only selections from that '07 draft who have panned out were the first two picks, forward Ethan Werek (June 7) and d-man Taylor Doherty (March 2). Doherty is coming off a disappointing second season, but will get drafted into the NHL because teams can't teach size (6-foot-8, 220 lbs. give or take).

Only two players on this past season's team, defenceman Brian Lashoff (July 16, 1990) and winger George Lovatsis (Nov. 30, 1989), were born in the second half of the year. Lashoff ended up being signed by the Detroit Red Wings after being passed over in the 2008 NHL draft, which makes him a late bloomer. Lovatsis proved to be a fan favourite last season. They were, on balance, decent additions.

As you might expect with a 19th-place hockey club, the Frontenacs make a lot of in-season trades. One of general mangler Larry Mavety's trademarks is to deal for a former high draft choice, telling anyone who will listen that he's bound to realize his potential with the Fronts.

This season the Frontenacs traded for three forwards who had each gone relatively high in the draft, Kelly Geoffrey, Colt Kennedy and Mitch Lebar. You know where this is going: Their birthdays fall within a five-day span in late January, the 25th, 28th and 29th. Perhaps it is true they haven't found the right fit. Based on the evidence, it might be that starting from an early age, they had a leg up on their peers. That made them seem better than they really were, and now their peer group has closed the gap.

Windsor has an abundance of players who didn't have the benefit of an early birthday and/or were not drafted early at age 16. They're the champs. Kingston, the Siberia of major junior hockey, has all these players with early birthdays who were taken high and are now perceived as not playing to potential. In the words of an old Twilight Zone episode, how many coincidences add up to a fact?

There are other factors in building a winner in the OHL. How players are treated, fan and community support, being able to bend the draft rules, each play a role. The best this analysis can do is be a gauge for who can find the hidden gem, and who doesn't know their ass from a hole in the ground.

(In the recent draft, the Frontenacs used the 37th and 42nd overall picks on players not born in January, February and March. Winger Brett Morgan, a late second-rounder, was born Aug. 12. Clark Seymour, the defenceman taken early in the third, was born May 18. There might be hope yet for this franchise!)


Marc Foster said...
This comment has been removed by the author.
Andrew Bucholtz said...

Very interesting stuff, Neate. I think Gladwell is absolutely right about there being a selection bias in favour of players with earlier birthdays. As you wrote though, you certainly can't draft based only on birthdays. The other thing is that thanks to the bias coming in so early in the system (10-11 year old rep hockey, when those extra months make a huge difference), it can affect development to a big degree. Thus, if two players have the exact same physical attributes but one is a January birthday and the other is a October birthday, the January player is probably a better choice as he's likely played against better competition and developed his skills more. Transcendant late-born talents still shine through, but the more average players get squeezed out.

If you have a willingness to work with more raw prospects, then you might be able to get some great talent from unpolished late-birthday guys with late draft picks. That seems to be along the lines of what Windsor's doing, and that makes a lot of sense; it's the idea of finding overlooked talent, which I'm always in favour of.

I'm not sure that a direct birthday comparison between Windsor and Kelowna proves too much, though; it's certainly interesting to look at, but at the end of the day, these are two of the best teams in Canadian junior hockey. They both wound up being very successful with starkly different birthdays of players on their roster. Just because Windsor won the last game doesn't mean there's much between the teams.

The Windsor - Kingston comparison is much more important in my mind. For one, it's a good team compared against a bad team. For another thing, the logic seems to make sense; one of the big criticisms of Mavety has been his draft record and ability to evaluate talent. Unimaginative talent evaluators tend to go with guys who are generally thought to be good by the scouting community, which is where the early-year bias comes in; early-year players might put up more stats at a time when later-year guys are only showing flashes of potential. It might be worth Mavety's while to look at some more unconventional players this year (and he'd probably have a better chance of getting them to come to town as well!).

Dennis Prouse said...

Windsor was the best team in junior hockey wire to wire this year, so it is fitting that they won the Memorial Cup. Still, that doesn't change the fact that the tournament format sucks. :-)

Anonymous said...

HOw would you like it to be.


Dennis Prouse said...

Neate and I have mused about a "Final Four" type format -- two Semi-Finals on Saturday, a Final on Monday Night. It won't happen, as the current format is a nice cash cow for CHL, but that would be ideal, cutting down on the number of meaningless games.

Here's an illustration of how bad the format is at present -- both Drummondville and Rimouski beat Windsor during the Round Robin, but they got to sit at home and watch while Windsor got to the Final and won. If all you need is a 1-2 record to make the playoffs, why bother even having a Round Robin?