Revenge of the Nerds
The Bill James of hockey? Columnist Greg Wyshynski talks to
the man who's trying to revolutionize the way the NHL keeps
stats and evaluates its best players.
(WASHINGTON, DC) -- Who is the most productive offensive
player in the NHL this season? The numbers tell us, as
of Nov. 15th, that it's Marian Hossa of the Atlanta
Thrashers with 15 goals and 30 points — leading the
league in both offensive categories.
But can we really trust beauty pageant stats to tell us
everything about a player's true value?
As hockey fans check the leader-boards while waiting to
see who will shine up this year's Art Ross or Maurice
Richard trophy, baseball sabermetricians are applying AP
Calculus theorems to offensive figures — mining
statistics for candid evaluations of a player's merit.
Chicks might still dig the long-ball, but baseball's
intelligentsia stopped treating RBI and slugging stats
as gospel the minute Oakland A's General Manager Billy
Beane accepted "Baseball Abstract" author Bill James as
his sabermetric savior in the mid-1990s.
I'm not ready to christen Gary Cohen as hockey's answer to
James — he admits he's not nearly the obsessive mathematician.
But Cohen may be hockey's first statistical revolutionary:
expanding the limited scope of current player data, and
establishing entirely new standards for measuring true on-ice
"[Baseball] goes really deep into analyzing statistics. Hockey
keeps it very basic — goals, assists, that sort of thing," he
said. "I still think there are a lot more stats that hockey
can use that tell you about a player."
Cohen is the man behind Hockey-Recap.com, a daily hockey fix
that's all about the numbers and what's behind them. He keeps
the basic stats, but includes a few that can't be readily
found. I used the word "revolutionary" before, and I meant it:
this guy's going to change the way we calculate, calibrate,
evaluate and disseminate stats in the NHL ... even if his site
looks like a no-frills version of the Drudge Report.
thesis is that hockey fans are being mislead by the
current numbers. He calls plus/minus "worthless,"
because so many players are given a role that
emphasizes defense over scoring. He thinks giveaways
and takeaways are as much a subjective mess as the
nebulous "hits" category.
"I've watched games and looked at the stats and I have
no idea how they came up with those numbers," he said.
(And he's got about as much love for goaltenders as
"There's no goalie stat that tells you anything,
unless [the league] starts breaking the ice down into
zones and rating shots that come at them."
It's that sort of statistical specificity that Cohen's
fighting for. Think of NHL stats like television ratings (I
know: sore subject, hockey fans). The newspaper lists the
broad categories: goals and points, ratings and shares. The
advertisers, however, only care about explicit demographic
information; broken down by gender, age, geography —even the
number of minutes they viewed the program before flipping over
to an old "Cheers" rerun.
Cohen yearns for that kind of detail. He wants first assists
tracked. He wants a nightly shift report — how much did Alex
Ovechkin play with the top line last game, and how much did he
play with other mismatched pairings? Some of these numbers he
needs the NHL to tabulate; others he plans on adding to his
site in the future, like shot-types per player, face-off
success by zone and other “ice-o-metrics” that include clutch
goals. In the end, like James did in baseball, Cohen hopes
fans will utilize all the numbers to evaluate and project a
player’s efficiency rather than relying on the limited
capacity of glamour stats.
For example: who did we say was the most productive offensive
player in the NHL this season? According to Cohen, it’s not
Hossa — he’s the sixth-most productive. No, the most
productive is Buffalo’s Maxim Afinogenov, thanks to a stat
Cohen created that measures the number of seconds between
points on the ice for a player. So while he has 11 points less
than Hossa as of Nov. 15th, Afinogenov (10:52) is still more
productive than the NHL’s “best” offensive player (14:25).
So how does Cohen manage to take a night’s worth of facts and
figures and turn them into a daily Web site and e-mail
The 33-year-old, who lives near Montreal, started with a
baseball site called TheBaseballCube.com, which culled both
Major and Minor League stats. He turned to hockey, his true
passion, this season. The process is fairly simple: he cuts
and pastes box scores from sites like Yahoo! Sports and
NHL.com into a database, and writes programs that sift through
and order them in different categories. Without enhancements —
like injury reports and a roll call of hockey stories from
around the Web — Cohen said the updates take about 20 minutes
He said he didn’t realize until he began his site that
arbitration and contract negotiations are two good reasons why
the NHL isn’t in the business of keeping a multitude of
different stats. But that could change if its members demand
otherwise: Cohen said teams like the Minnesota Wild are
petitioning to get the league to collect more game-day data in
order to better evaluate their own players.
“It’s really hard to come up with new stats because it’s a
free-flowing game. It’s not like baseball; things are always
moving in hockey,” said Cohen.
But baseball and hockey do share one thing: sacred records.
And just like the home run has become about as commonplace as
a Rick DiPietro joke, Cohen believes that statistic benchmarks
in the NHL could become just as pedestrian under these new
offensively emphatic rules.
“That’s one of the things I’ve had against baseball.
[Milestones] were relevant for a while. To tell you the truth,
I don’t care so much about the steroids as I do that the
numbers are all different now. You have to have that benchmark
where you know what’s good. Fifty [goals] needs to be good; 30
home runs need to be 30 home runs,” he said.
“[The NHL will] have to bring it back the other way to
defense. I think that’s what they want to do. I don’t think it
was so much that the game was boring [in the 1990s]; it’s that
there were no goals. I think they want to keep the stats where
they mean something, so that they mean something from
generation to generation.”
Even for stat-obsessed American sports fans.
Soccer doesn’t work in the United States because there’s no
scoring, just like basketball thrives here because it’s all
scoring. Baseball is passed down through the generations
because it’s as much about stats and speculation as it is
slugging and strikeouts. American fans live to quantify, from
the number-crunching on fantasy league Web sites to the pages
of sports record books.
But Cohen doesn’t believe that casual fans are interested in
“I’m not even sure that when Wayne Gretzky was scoring 80 or
90 goals that it was even grabbing attention in the States,”
he said. “I just don’t think they understand the game enough.”
So even if scoring records started falling more quickly than
the Flyers’ playoff hopes, Cohen doesn’t believe American
sports fans would care — unless you’re talking about the ones
with a copy of “Moneyball” on their bookshelves.
“If they could study a game, study the stats that really tell
the story…I think the intelligent types that really like that
sort of thing would be more into hockey.”
Like I said, it’s a revolutionary notion: that the key to
drawing new fans to the NHL isn’t in catering to Neanderthalic
predilections for goal-scoring and the occasional fisticuffs,
but in finding an access point for intellectual sports
aficionados to begin devouring hockey stats. We’ve spent the
last 20 years marketing to Bill Murray when it's Bill Gates
who might actually end up appreciating the game. Forget pink
jerseys for the ladies; where’s the hockey swag for the nerds?
Officially licensed pocket protectors! NHLPA graphing
calculators! Can't you hear it now? "Hi, I’'m Marty Turco for
OK, perhaps that’s a bit of a reach. Then again, I'm sure
those who read the first "Bill James Baseball Abstract" in
1977 didn’t think its theories would one day become
mainstream, and that its author would be hired by the Boston
Red Sox to help mastermind their first World Championship in
So stranger things have happened; just don't shine up that
Maxim Afinogenov Trophy for Best Production quite yet.
"Mike Ribiero would probably win it [someday]," said Cohen of
the former Hab, "and I wouldn't want that to happen."
Wyshynski, also the Sports Editor of The Connect Newspaper, is
a columnist for TheFourthPeriod.com and the Washington
Correspondent for The Fourth Period Magazine.
His book, "Glow Pucks and 10-Cent Beer: The 101 Worst Ideas in Sports
is now on sale.