Does anyone miss hockey analysis before analytics?

[Matty Sundin]

I miss when they had more segments where they actually dissect and describe a play piece by piece what is happening. McGuire used to do that before he went full moron.

Even still I genuinely don’t care if someone has only hit the net 3 times in his last 10 shots attempts. I’ll look it up online if care that much.

 

[Legionnaire11]

Not at all.

Doesn't anybody remember what passed for analysis before analytics?

"They need the WILL TO WIN! They just gotta WANT the puck more!"

"Need more #grit, that's why they're losing"

"Gotta bear down, and put those chances in the net"

It was all narrative driven garbage.

Even if what you're saying is 100% true (it's not) it's perfectly fine for what fans watching at home need. In fact, I'd say building up that emotional aspect and romanticism of the game is better for fostering passion amongst viewers that links them to the game rather than throwing a bunch of numbers at them.

 

[Mattilaus]

It couldn't be because they got 4 easy wins against AHL teams in Edmonton and San Jose inflating things?

So what you are saying, is that they will eventually have to face good teams, and it will go down....or regress?

 

[tucker3434]

Advanced stats were definitely alive and well in 13-14. That's all people talked about when the Avs were winning that year. And they were ultimately right.

Through it all, you can come down to a simple conclusion, teams that get more high quality opportunities than their opponents are more likely to win consistently. Shot differentials alone can get you in the ballpark, but it definitely doesn't hurt to have an idea of quality chances versus the teams that are just throwing pucks on net from the blue line.

 

[Mattilaus]

Advanced stats were definitely alive and well in 13-14. That's all people talked about when the Avs were winning that year. And they were ultimately right.

Through it all, you can come down to a simple conclusion, teams that get more high quality opportunities than their opponents are more likely to win consistently. Shot differentials alone can get you in the ballpark, but it definitely doesn't hurt to have an idea of quality chances versus the teams that are just throwing pucks on net from the blue line.

Ahhh I see you have been watching the Sabres this year.

 

[zappa4ever]

Fancy Stats = meh. (mainly pointing to compiled data stats)

What's funny is how they come up with the fancy stat by:
1) choosing each factor
and
2) how they're weighted

go look up the source material these developers are using and it becomes quite a stretch to reduce it to:
Player A = teh aawesomez!
Player B = suxz

using 20 or so data points for something like xGF make me want to vomit in terror[/homer]

Raw data is another story... a stat like O/Neutral/D zone starts is a single data point that has some value


LOL at stats like the Deserve to Win O-Meter and the Game Score Impact Cards (the 3 shades of blue for each player graph)

 

[JetsUK]

Suspect that the amplification of stats-based analysis in programming is also driven by the NHL and broadcasters' affiliation with the betting industry.

Numbers help make assessing odds accurately seem more plausible to some (many?) punters. We've seen it in racing form sheets for decades, and while the "analytics turn" is a thing. I suspect that its rise among a viewership allows those viewers to a) have reasons to be "right about" some key elements of a player or team's performance, and b) for betting shops to gain credibility for fairer odds and lucrative potential by offering information that helps to make informed netting possible.

Not claiming that these audiences are the same, but having watched footie analytics being effortlessly integrated into online betting in the UK/ Europe I'm not surprised to see hockey heading the same way.

My (fair odds) 2p.

 

[hohosaregood]

I think that analytics helped bring in a wave of hockey analysts that I appreciate having now. I feel like in the early 2010s, there just weren't that many outlets worth reading or interacting with. Like the deepest analysis I can remember that was worth anything was Justin Bourne's system breakdowns for The Score.

I don't really care for analysis work that revolves solely around analytics but if it's stuff that's like here are the analytics and here's the breakdown of why that's happening, I love that. I'd argue that the best analysts today blend the fancy stats and the breakdowns really well.

 

[Oleksiak]

So what you are saying, is that they will eventually have to face good teams, and it will go down....or regress?

It will, but you don't need trash produced by some geek with zero understanding of the sport to realize it. The sooner networks and teams realize that these spreadsheet heroes aren't qualified to make hockey decisions or give informed commentary, the better.

 

[Bear of Bad News]

It will, but you don't need trash produced by some geek with zero understand of the sport to realize it. The sooner networks and teams realize that these spreadsheet heroes aren't qualified to make hockey decisions or give informed commentary, the better.

You sound objective about the whole thing.

 

[Mattilaus]

It will, but you don't need trash produced by some geek with zero understand of the sport to realize it. The sooner networks and teams realize that these spreadsheet heroes aren't qualified to make hockey decisions or give informed commentary, the better.

Why the insults?

 

[SniperOnTheWing]

Yes I’m sure the Canucks will continue to face below league average goaltending from every opponent they face all season long. Only 67 more games to go!

 

[Roo Returns]

The Colorado Avalanche literally built the bulk of their team using analytics

That makes no sense when Mackinnon, Rantanen, and Makar were all top picks. But ok.

 

[Bear of Bad News]

That makes no sense when Mackinnon, Rantanen, and Makar were all top picks. But ok.

If it were that easy, I can think of other teams who’d love to do it.

 

[Mattilaus]

If it were that easy, I can think of other teams who’d love to do it.

I feel personally attacked

 

[JaegerDice]

That makes no sense when Mackinnon, Rantanen, and Makar were all top picks. But ok.

Wow, I had no idea the Avalanche iced only 3 players during their contending seasons. You sure did show me.

 

[MrLouniverse]

It's the difference between the old Travel Channel World Poker Tour episodes where people would wear capes and taunt their opponents with clown props while the announcers talked about putting a man on a hand, and PokerGO streams where they talk about 3betting frequencies and range advantage.

 

[Roo Returns]

Wow, I had no idea the Avalanche iced only 3 players during their contending seasons. You sure did show me.

Geez Louise, they didn't just plug a regression model into a computer with 7 variables and select players from an entry draft, or do the same at the trade deadline.

There were a number of factors that went into all of their personnel decisions. I never said analytics were not involved. They weren't the only decision. Personality, on ice systems, salary, age, etc.

There's a lot that goes into this.

The "sole analytics" champions make it seem like some undergrad with nine credits in Statistics who taught themselves R found all 7th rounders in some tier two European League and won a Cup. And high five themselves in the process.

This is contrarianism at its finest. If the internet isn't full of rage and shaming, then it's contrarianism. You can say humans should eat vegetables and drink water someone is going to beat their chest and say otherwise.

So yes, the Avalanche used some analytics. They didn't have Skynet as their GM.

 

[JaegerDice]

Geez Louise, they didn't just plug a regression model into a computer with 7 variables and select players from an entry draft, or do the same at the trade deadline.

There were a number of factors that went into all of their personnel decisions. I never said analytics were not involved. They weren't the only decision. Personality, on ice systems, salary, age, etc.

There's a lot that goes into this.

The "sole analytics" champions make it seem like some undergrad with nine credits in Statistics who taught themselves R found all 7th rounders in some tier two European League and won a Cup. And high five themselves in the process.

This is contrarianism at its finest. If the internet isn't full of rage and shaming, then it's contrarianism. You can say humans should eat vegetables and drink water someone is going to beat their chest and say otherwise.

So yes, the Avalanche used some analytics. They didn't have Skynet as their GM.

After the fiasco of the Roy years, the ownership of the Avalanche went to Sakic and told him the Duchene trade had to be handled differently because it was so crucial. No more ‘good hockey men’ back-patting their buddies.

The Avalanche already employed Arik Parnass, one of the godfathers of hockey analytics. He was empowered and his division was expanded to integrate pro scouting and interface with amateur scouting. If they were going to get a package of picks and prospects, they werent going to waste them based solely on some scouts eye test.

They hired the likes of DTMAboutHeart and others straight off of twitter and hockey graphs. They were given point on every trade and FA signing moving forward. No move was made without their blessing. Sakic handled contracts and relationships. Every acquisition, from Nichuskin to Devon Toews, was made on the basis of their valuation.

Kadri was a happy accident as they werent expecting him to go up for trade and when he did, they were expected to be outbid. They were, by Calgary, but Kadri nixed the trade. They were the beneficiaries. So you can give Sakic credit for the relationship with Dubas on that one.

That said, of ALL the teams in the league, Colorado and Carolina are the two most analytics driven teams.

 

[D Wakaluk]

You could say the more biased fans have become more annoying with the help of advanced stats.

Nothing wrong with more data but it absolutely transforms into crazy juice for some

 

[Rejavanated]

No. I think analysis without analytics is mostly performance without substance.

I'm way more bothered by the infestation of betting shit in the broadcast than conversations about advanced stats.

 

[80shockeywasbuns]

Not at all.

Doesn't anybody remember what passed for analysis before analytics?

"They need the WILL TO WIN! They just gotta WANT the puck more!"

"Need more #grit, that's why they're losing"

"Gotta bear down, and put those chances in the net"

It was all narrative driven garbage.

It’s all inane nonsense. The idea that “compete level” or “will to win” are things that can separate pro hockey teams is the sports fan equivalent of believing in Santa.

Personally I think analytics are great for sports because they drive out a lot of these old man tropes/philosophies and modernize the type of play that teams prioritize. The quality of hockey will only improve as analytics help to reveal the uselessness of hitting, size, fighting, faceoffs, shot blocking, “cup pedigree”, etc

It just seems like analytics make a lot of fans uncomfortable because they feel like sports talk is just supposed to be dudes shooting the shit, parroting cliches, saying whatever low effort takes come to mind. I can totally understand how people can find WAR models annoying but simple differential stats like xGF%, corsi, and scoring chances should all be mainstream stats that are incorporated in a broadcast. There is nothing advanced about any of these things, its time to scrap the faceoff% and +/- graphics

 

[ijuka]

It’s all inane nonsense. The idea that “compete level” or “will to win” are things that can separate pro hockey teams is the sports fan equivalent of believing in Santa.

Personally I think analytics are great for sports because they drive out a lot of these old man tropes/philosophies and modernize the type of play that teams prioritize. The quality of hockey will only improve as analytics help to reveal the uselessness of hitting, size, faceoffs, shot blocking, “cup pedigree”, etc

It just seems like analytics make a lot of fans uncomfortable because they feel like sports talk is just supposed to be dudes shooting the shit, parroting cliches, saying whatever low effort takes come to mind. I can totally understand how people can find WAR models annoying but simple differential stats like xGF%, corsi, and scoring chances should all be mainstream stats that are incorporated in a broadcast. There is nothing advanced about any of these things, its time to scrap the faceoff% and +/- graphics

Corsi's the only objective stat out of those. xGF is subjective, and hasn't even been calibrated accurately historically. Scoring chances are subjective, as are high-danger scoring chances. Different sources are going to have different definitions.

Also, a high-danger chance by eyes might be quite different from what analytics that don't actually see the games consider high-danger chances. Those could well be completely harmless due to some factor that you can see with your eyes, but that the spreadsheets cannot capture.

 

[Colorado Avalanche]

The problem with analytics are the people who use them, not the analytics themselves. Just like any sort of statistical data set.

"Analytics" aren't this magical thing, they're basically just shot differentials. That's it, any sort of "advanced stat" is just a measurement of shots. The people who overuse them think they're gospel and the people who hate them think they're nonsense, but both sides are wrong. Analytics are a supporting point to an argument, not the entire argument. It's the same thing as the eye test or basic stats.

I think the largest failure of "analytics" was calling them "advanced stats", because they are not "advanced stats" whatsoever. The reality is that "advanced stats" are just some measurement of shot differentials, that's it. They're effectively +/- for shots.

CF is just a shot attempt and CF% just shows what percentage of shot attempts your team is taking. FF is just unblocked shot attempts and FF% just shows what percentage of unblocked shot attempts your team is taking. xGF is just shot attempts multiplied by a location probability factor that converts that shot attempt in that specific location into an "expected goal".

None of these stats are these magical stats, both from the arguments of "they're the truth" or "they're made up".

Great post

 

[Channelcat]

I'm not hearing analytics on the broadcasts.