Colorado Avalanche: Is Roy's Stance on Advanced Stats Uncommon?

3 of 3

Oct 14, 2015; Denver, CO, USA; Colorado Avalanche head coach Patrick Roy reacts during the second period against the Boston Bruins at Pepsi Center. Mandatory Credit: Chris Humphreys-USA TODAY Sports

Colorado Avalanche head coach Patrick Roy is drawing heat for his remarks on hockey’s fascination with advanced stats. Is his opinion really that off-the-wall?

Third year head coach and former player-hero Patrick Roy has found himself in the hot seat with Colorado Avalanche fans and media after his October 21st press conference following the teams’ practice. Looking mildly irritated with the media whom have been questioning his every decision as of late, Roy found himself having to defend the teams’ statistical poor showing that has accompanied their season. Most notably – their Corsi and Fenwick %’s.

“We’ve been looking at all the games. And obviously if you’re looking at our Corsi or Fen[wick], our numbers are not very good. I don’t think it’s because of the number of shots we’ve been getting, it’s more the shots we’re not taking. For instance if you’re looking at Corsi, the part I don’t like about the Corsi is you could shoot from the red line or from a terrible angle, and your Corsi will look good. Puck possession has nothing to do with it. Fenwick… there’s a bit of puck possession in there, but same principle. It’s more like shot attempt. If a guy shoots from the red line and it’s blocked, it’s still a shot attempt. This is something we don’t do very well, or we don’t think about doing a lot.”

Mile High Sticking Editor Janik Beichler made some excellent points in dissecting the meaning of Roy’s comments and how both Roy’s comments and the statistical data itself can and is being taken out of context. Advanced statistics, or “fancy stats” as they are often referred to, have been gaining popularity in hockey for several seasons now, prompting hockey fans, critics and the organizations themselves to take heed of their interpretation.

More from Mile High Sticking

For the uninitiated, Corsi and Fenwick are the names given to statistics that track a teams’ shots on goal. Jim Corsi, the current goaltending coach of the St. Louis Blues, came up with the idea of tracking all shots on goal as a way to gauge how active a goalie was in a contest. Matt Fenwick, a Calgary Flames blogger modified the stat to a minor degree by tracking only unblocked shot attempts.

They are often referenced in regards to puck possession, which is a bit confusing, but mostly as a way to say that – if you are shooting more than the other team, your possession of the puck must be higher as well. The Hockey Writers did an excellent breakdown of how this is all actually figured, which illustrates why these are important stats to be familiar with.

If you’re feeling confused about what makes these figures so “advanced” – you’re not alone. Even the Hockey Writers admit it’s a bit of a misnomer. In my opinion, Corsi and Fenwick are a more detailed look at the old “shots on goal” tally with an emphasis on differentials between what the other team is doing and how the team has done against several teams over a longer period of time.

What happens, advanced statisticians will argue, is that you begin to see trends emerge. The point the media is trying to make, simply, is that the Avalanche are getting badly outshot from several teams over a long period of time. And they’re really not wrong if you look at the numbers. As head coach Roy tried to elucidate to the media in his conference however, being outshot is a symptom of the problem – not the problem itself.

So, it all starts to get rather sticky pretty quickly. As Roy pointed out, the team could “remedy” the statistical problem by taking more shots from low-percentage areas. The results, he would probably argue, is now that you’ve got great Corsi and Fenwick numbers, but you still haven’t fixed what’s broken. While he’s technically wrong about Fenwick and blocked vs. unblocked shot attempts, the overall argument he’s making is still valid.

The temptation in the hockey world is to say that you can mostly tell who’s a good team and who isn’t by looking at their puck possession, which is factored by advanced stats like Corsi and Fenwick %’s. For many players and coaches, this simple identification makes them bristle. Patrick Roy is certainly not the first former player to take contention with the analysis, and to make things interesting – he’s not even the most vocal opponent of them.

So who, you might be asking, is more passionate about the need to take the emphasis off of percentages and formulas to determine talent in professional sports?

Next: Analytics Most Vocal Detractor

May 20, 2015; Atlanta, GA, USA; Former NBA player and current TNT television personality Charles Barkley prior to game one of the Eastern Conference Finals of the NBA Playoffs between the Atlanta Hawks and the Cleveland Cavaliers at Philips Arena. Mandatory Credit: Brett Davis-USA TODAY Sports

Former NBA star Charles Barkley is probably one of the most vocal opponents of using analytics as a measuring stick for team success that you’ll find. Similar to the NHL, the NBA has also tried to mimic the success of advanced stats used in baseball during the “Moneyball” era. In baseball, with a larger number of moving pieces determining success, this can be of incredible use in determining player deployment.

Barkley believes that determining what makes up a good team is not decided by numbers, but by talent. As a former player, he argues, his ability to distinguish between good and bad play is more keenly honed than any statistician could attempt to emulate.

“The NBA is about talent. All these guys who run these organizations who talk about analytics, they have one thing in common — they’re a bunch of guys who have never played the game, and they never got the girls in high school, and they just want to get in the game… Analytics don’t work at all. It’s just some crap some people who are really smart made up to try to get in the game because they had no talent.” – Charles Barkley

While Barkley’s opinion was a bit vitriolic, he does make a point. A few seasons back, when the NBA analytics crowd swooned over the Houston Rockets’ advanced statistics (which favored their ability to limit opponent’s shots from prime areas), Barkley argued otherwise.

“Just because you’ve got good stats doesn’t mean you’re a good team defensively. They’re not a good defensive team. [The Rockets] give up 118 points – no good team gives up 118 points. What analytics did the Miami Heat [have]? What analytics did the Bulls have? What analytics do the Spurs have? They have the best players. They have coaching staffs who make players better…I’ve always believed analytics was crap, and you know I never mention the Rockets as a legitimate contender, because they’re not.”

The way I interpret what Barkley was trying to get at is this: former players (like Patrick Roy) have been able to see what it takes to win games from a perspective that fans and analysts don’t have. What an advanced stat line will compute can be interpreted out-of-context to the overall themes of the game.

They can’t account for many of the intangibles. For instance: how would advanced stats line interpet a teams’ dip in performance against the Nuggets while playing at altitude that they normally aren’t subjected to? How about if their star player is hurt and missing from the roster? Barkley, like Roy in his comments, devalue the statistics then because they don’t tell the whole story. As a coach, Roy is privy to more of the calculation than just the numbers.

To a degree, I completely agree with Roy and Barkley. As Roy pointed out, shooting from bad areas to improve the look of your statistics is most likely not going to help you win games. Therefore, if the criteria of the stats don’t take in certain fluid complexities to the game, then why take them seriously?

On the other hand, it’s hard to argue that advanced statistics aren’t indicative of a greater overall trend if the sample size is adequate. Over the course of one or more seasons, the advanced stats start to tell a story when compared to other teams around the league. The story they tell about the Colorado Avalanche has not been good for several seasons now.

When the Avalanche came crashing back down to Earth last season following a dream season in 2013-2014, the analytics crowd was there to say, “I told you so.”

Next: Reaching a Middle Ground?

Sep 24, 2015; Denver, CO, USA; Colorado Avalanche center Marc-Andre Cliche (24) takes a shot on the goal of the Calgary Flames in the second period during a preseason game at Pepsi Center. Mandatory Credit: Ron Chenoy-USA TODAY Sports

Full disclosure: I am not as smart as Patrick Roy. I don’t watch the team practice every day. I don’t have access to team meetings. I don’t review film with the coaches. I’ve never even had a conversation with anyone on the team. That does not qualify me to pretend that I know more about the team than Roy or the assistant coaches do. It doesn’t stop me from having my own opinion on the Avalanche, which is in part, formed around their advanced statistics numbers, however.

I think what Roy and Barkley were trying to get at in their comments is that advanced statistics are a good tool to use, but not the final product on whether a team is good or bad. If I were building a wooden chair, I would need to use several tools during different stages of the construction. If I attempted to use only one tool at my disposal for the whole operation, let’s say a hammer, I would yield a pretty awful chair. The same goes for coaching a hockey team.

While I certainly don’t discount the importance of analytics, I think it would be asinine as a fan to think that you can coach a team to success using only analytics or that analytics is the end-all-be-all determinate of success. The harsh nature of statistics to either be “good” or “bad” is a bit too simple in the NHL – now or ever. Context is always needed and several factors interplay to affect others.

Neither do I think that the Avalanche’s relatively poor Corsi and Fenwick numbers are to be taken lightly. I tend to think of them as the canary in the coal mine for the Avs. Something is amiss in the Avalanche’s game and the results are filtering through to their advanced stats. Roy points out that these stats can be cheated around by taking more shots from rubbish areas, but it would behoove him, in my opinion, to acknowledge that the stats are indicative of something much larger that’s gone afoot.

Jeff Van Gundy, the head coach of the aforementioned Houston Rockets who did compile his roster based on his statistical findings and found success doing so added this in response to Barkley’s comments on a panel at the MIT Sloan Sports Analytics Conference.

“It’s like we have a church of analytics and you have to be a devoted follower,” he says, to laughter. “You can believe in math, but also intangibles, he says. “I hear people here say things like, ‘Oh, you actually believe in hard work?” -Jeff Van Gundy

Tyler Dellow, a hockey researcher recently hired by the Edmonton Oilers sympathizes with the resistance to the analytics movement. Regarding players and coaches interpreting the data, he put it like this:

“You have people who don’t speak the same language. They may not be very interested in your graph or your scatter plot…There’s a saying in British soccer that the table doesn’t lie. Well, it doesn’t in that the numbers are added up correctly, but it does in terms of telling you who is good.” – Tyler Dellow

At the end of the day, Patrick Roy is paid to help the team obtain favorable results. Lots of teams struggle with puck possession and limiting other teams’ scoring chances, but when losses start to pile up, the odor of these sorts of things begin to ripen. Winning, above all else, will serve as the best deodorant for Roy’s stats detractors. As fans, let’s try not to miss the forest for the trees. Roy is making adjustments and we should all avoid jumping to conclusions prematurely.

Next: Avs Offense Shouldn't Rely on Rushes

More from Mile High Sticking

Home / Avalanche News