One of the things I’ve long considered interesting in the analytics vs traditional analysis debate is that what analytics point out isn’t that new. In fact the ‘traditionalists’ may have long been speaking analytically but never understood how they were doing so. This is not some call to arms to justify one side or the other in the long standing debate, it’s an attempt to bring the sides closer together. “Look guys we’re not that different, we both want the same thing, better hockey”.

How do I see this bridging of the great divide? Taking a look at how traditionalist hockey thinking is actually very similar to what analytics are now telling us. What gets lost in the debate is that analytics are quantifying what coaches want via strategy and tactics. Analytics are measuring the traditionalist thinking and helping assign a more accurate value to what goes in a hockey game.

Basically, through data gathering, the great folks who provide much of this analytical information are producing a value for players who do the things coaches want done. So without further explanation, here is how stats and traditionalists are closer together than they know.



A traditionalist might say… “Good things happen when you go to the net”
Analytics might say.. “this player has a great success rate in the high danger scoring area”

The idea here is that the closer you get to the net the goalie has a harder time reacting to the shot attempt. Look at there the bulk of goals are scored in the NHL, it’s in the front of the crease and that’s the high danger area, you don’t want players getting the puck there because they have a far better chance of scoring.

A traditionalist might say…”A 3-1 lead is the worst lead in hockey”
Analytics might say…”Score effects can present to challenges holding a lead”

It’s pretty simple really, teams do things to get a lead and once achieving the lead stop doing the things that got them the lead. So in the end this is what ‘score effects’ are; the effect the score has on the way the game is played. The leading team usually starts giving up more shots and the trailing teams starts pushing more and taking more shots.

A traditionalist might say… “we need some puck luck, or the bounces didn’t go our way”
Analytics might say… “PDO, the regression to the average of historical shooting percentage added to save percentage which should approach 1000”

It does not so much as measure luck but measures the variances in things that might involve bounces and if the are favourable or not. Basically what we know is that very few players shoot higher than average for extended periods and the same with goalies and save percentage. Some of those numbers can be based around variables and chance.. or good and bad luck.

A traditionalist might say… “You can’t score if you don’t shoot the puck”.
Analytics might say… “Shot frequency is the good indicator to future goal scoring”.

Basically what the vast volumes of work with analytics folks has shown is that measuring the more frequent events helps predict other less frequent events, like goal scoring. So if you are not attempting to shoot the puck, it’s never going to cross the goal line- simple right. Well if you are attempting to take lots of shots you have a better chance of scoring goals in the long run.

A traditionalist might say…“The key to good offence is good defence”
Analytics might say…”A defence man who can transition the puck out of the zone is the start to good offence”

This phrase is a great one because it really touches to the core of some great work about zones, possession and player roles. If the player can stop shots from getting through that’s fine but if he can’t get the puck out of his zone we already know that could lead to more shots against and possibly more goals. It’s the same idea as offence starts with the defence and that’s why we have old terms like ‘puck-moving defence man’.


A traditionalist might say… “We’ve got to score dirty goals” or “Need to get more traffic in front of the net”.
Analytics might say…”You can have greater goal scoring success from the high danger areas of the ice”

Sure the high danger area point has been covered but let’s reinforce it a bit more. Dirty goals and traffic happen right where it’s hardest for the goalie to cover and react. By getting the puck there and shots you are giving yourself a far better chance at success and the data proves this time and time again.

A traditionalist might say…”You have to play a 200 foot game”.
Analytics might say…”Zone control and tactics are important to success”

This is another great cliche that has importance for both sides. The work being done on zone entries, exits, and transitions is shaping how we look at player roles and abilities. Guys like Bergeron, and others are true 200’ players and help drive play everywhere on the ice for their teams. What analytics help do is quantify which players do it best.

A traditionalist might say…”You build a team from the net out”
Analytics might say… “Are you freaking nuts goalies are voodoo!”

There has to be one contrarian thought in all of this and as far as I’m concerned this is the best one to use. The thing about goalies is you don’t know if you have Carey Price or Tuuka Rask or Lundqvist until you have them. Look at the history of goalies and there’s coaches and GMs who have all looked like geniuses or morons depending on the outcome of how their goaltending moves worked out. Devan Dubnyk could have been had for free by many, then there’s Bryzgalov. What about the hype surrounding ‘the Monster’ or where the hell did Tim Thomas come from?

A traditionalist might say… “Finish your checks”
Analytics might say…”Why use players who cant gain possession”

Okay I lied, another contrarian term. Finishing the check has long been held as a valuable trait for players because it means they play a hard game but think about it for a second. The purpose of checking is to separate your opponent from the puck so that you can gain posession of it. If you have to ‘finish’ your check it’s likely you got their too late, and the puck is already gone, but hey crash into the guy and take yourself out of the play too instead of getting into position. Hits are the one thing in hockey that look great but really only have distinct value when the hitter gains puck possession. Everyone loves big hits, that’s entertainment value for sure but it should not be at expense of acquiring possession.

A traditionalist might say…”He’s good at mucking it up along the boards”
Analytics might say…”This player could be profficient at puck recovery”

One of the things that has happened in the hockey data movement is the tracking of micro-stats, or events to create micro stats. So when a players mucks it up along the boards that could be seen as a micro-stat event depending on how some one wants to look at it. If that player’s team comes away with the puck then something positive happened but if they lose it and surrender a scoring chance then it’s a negative event. Some players like Mathieu Perrreault are very good at puck recovery because they can go in and muck around in the corners and come up with the puck frequently. In the offensive zone getting deep to pressure and recover pucks is very valuable teams place value on this by tracking it.


There are probably a few cliches I’ve missed here but I think you can see what the point is- both sides are saying the same thing, but with different phrases and terminology. That’s a good thing and a reason to be optimistic about what we understand about hockey.

Here’s where the discussion becomes a little bit difficult, the traditionalists will use the long-held phrases but the analytics might say, “nah, that’s not the right guy”.

Before we all have our heads explode let’s get down to the key difference here… analytics are helping identify who does these long-held valuable hockey ideas better, or worse, than others. Furthermore, the captured data says some of the traditional hockey ideas and tactics that have been preached as valuable might not be that valuable. I know right, mind blown.

But it’s not that bad. Really it isn’t. You can still enjoy hockey without having to worry about the new, changing valuation of skills. You don’t have to get into huge social media battles to prove something that you believe because you really do believe the same thing most of the time.

Let’s look at one more cliche.

A traditionalist might say “That guy knows how to use the body”.
Analytics might say, “That could be good or bad, let’s figure it out”.

If a player has a large amount of hits in his stats lines it raises some questions such as this one: is he running around chasing the play and only finishing his checks and missing the puck completely? If so, it probably means he’s hurting his team because not only is he hitting guys who have moved the puck away from them, that same player is now way out of the play and hardly useful. However, if that particular player used his body to separate an opposition player from the puck and had a lot of hits and a decent possession score, there might be something to further look at and see if indeed he was especially adept.

The point is that traditionalists do want the same thing as what analytics are saying as valuable. It’s that sometimes the players a traditionalist likes are not seen the same when data from their play is analyzed. Perhaps those who just like to watch games and make some observations about what they like could be more open to new information. To the contrary when the analytics present that a certain well-liked player is actually a train wreck, those citing the analytics should redirect their scorn. “Oh you like ‘so-and-so’ how about you watch this player more closely, he’s way better at doing the things you like”.

For you traditionalists instead of blurting out scorn towards some one citing data, ask a question. “Hey, I love dirty goals, who’s good at getting those?”

The fact is the change has come and it’s not going to stop. We went from cheques, to debit cards, to e-transfers to bitcoin. We moved from bankers hours to 24/7 phone apps. This level of convenience didn’t happen because everyone resisted change. That being said you can still visit your bank branch and do things the way you always have.

Sports is like this too. You can still enjoy the things you like but you might discover a few other things that make the games more enjoyable and interesting. Is that such a bad thing? There’s bound to be things either side dislikes but is it worth being hostile about it?

Take a recent Winnipeg Jets player example. For years there’s been a wide divide one particular player. It’s one player who has never really been highly influential in games or his play but a lightning rod for volatile discussion, Alexander Burmistrov. His first years in Winnipeg he was an analytics darling, but the last two have seen both traditionalists and analytics come together to generally see that Burmistrov is not the player anyone though he was. Maybe the traditionalists were ‘right’ earlier or perhaps the player has caught up to to where some thought he would be- a 4th line role player. The interesting part is that both sides have come to the same point using data and traditional methods of valuation.

Again, there’s more in common than there is in differences. It’s time to put the pitchforks and torches down and begin enjoying the sport for the very same reasons, smart effective hockey plays that help produce goals and wins. If we can’t even agree about that then it’s time to take a break from the sport, get some perspective and come back to enjoy it for what it really is, entertainment. We all have the right to find entertainment in hockey on our own terms. Let’s just remember that our ‘terms’ are not all that different and we don’t need to shove them down some one else’s throat.

Join the Discussion: » 7 Comments » Post New Comment

» Who are the Jets and time for me to let go...

» Jets bet Oilers in scoreless but exciting game.

» Digestion Problems: Jets edition

» Laine shines in 5-2 win over Wild

» Hellebuyck, Laine and Defense shine in 4-1 win over Penguins