Jump to content

Advanced Stats Bluff?


Recommended Posts

Without discussing "advanced" stats themselves, i'd like to ask a question:

Does anyone else believe that the hirings of Tyler Dellow and Kyle Dubas are a PR bluff? Sunny Mehta was hired by the Devils, but to me that makes sense because of the attention to detail that franchise has.

If you don't believe that a franchise would hire executives at least in part because of PR-put on your Habs jersey and sit in front of a mirror.

Makes perfect sense to me. The media in Toronto has been embroilded in shot attempt fever. PPP foolishly cried out in joy after the hire of Dubas, although the hype seems to have amounted to the signing of Daniel Winnik.

Katz and Lowe would probably hire Asimo if he could keep their fans from throwing jerseys on the ice.

I predict with his awful disposition and the Old Boys' Club in Edmonton that Dellow will be fired within months. Maybe Dubas can do some good if he doesn't choke on MLSE hubris.

The real challenge for these guys will be Carlyle and trying to rearrange furniture in Eakin's Playhouse.

Link to comment
Share on other sites

I don't think the hirings are intended to be a PR stunt. Advanced (new) stats have come a long way in recent years and teams, if they haven't already, need to make sure they have someone that knows their way around them to keep up with the rest. Every team needed a capologist but some waited a while before bringing one in, eventually everyone went and got one. It's a management position and teams often make announcements when someone gets hired for a management position. The Leafs may have went a bit far with the impromptu presser for an assistant GM but at the same time, everyone knew it was going to draw some attention so getting it over with early probably was the smart play.

Link to comment
Share on other sites

Dubas is more than just a stats guy. He just got that reputation because he follows them and is relatively young. The guy has a very good reputation which some of the more snarky Toronto reporters decided to ignore or rather were completely in the dark on because they are not good at their jobs.

Dubas has more in common with Marc Crawford (young guy who rose fast) than say Rob Vollman.

Link to comment
Share on other sites

Don't care much bout advanced stats (for baseball nor NHL) other than being sick of hearing about its importance, BUT, sure hope Habs mgmt uses it and every other thing they can think of to gain an advantage.

I wasn't a fan of stats courses and found it a frickin mind-numbing subject.

And I would bet that the vast majority of ball/hockey fans are fine with using results like, standings, win/losses, player production #s, etc and could care less to dig much deeper than that.

Link to comment
Share on other sites

I think it's all BS. Sure, Dubas did a smashing job getting out of the first round with SOO, but he was replacing GMs, maligned or not, that had double the experience he did. Management of any kind is all about experience. People come into management, sports or otherwise, with the idea that they can reinvent the wheel and get beat down by the system.

Shanahan in particular is very PR saavy. The biggest success of his tenure as the suspensions guy were his videos.

Blogs are a threat to franchises because they can't be controlled like the MSM. For me, they hired these guys at least in part to shut the fans up.

Dubas and I are the same age. I'm punching above my weight trying to run a 70 agent call center with a supervisor under me. This guy is supposed to run the Toronto Maple Leafs?!?!?! This isn't a restaurant, where a 14 year old dishwasher can be an executive chef in 10 years.

Link to comment
Share on other sites

Dubas and I are the same age. I'm punching above my weight trying to run a 70 agent call center with a supervisor under me. This guy is supposed to run the Toronto Maple Leafs?!?!?! This isn't a restaurant, where a 14 year old dishwasher can be an executive chef in 10 years.

In all fairness, Dubas has shown the drive and ethic to perform well above his 'weight class' by working his ass off in a league that paid him peanuts, and doing it very successfully at that. I do not think it is even remotely fair to compare your situation to his.

I think MAYBE a very small part of it is PR, but on the other hand he has shown to be a very savvy manager who also understands and utilizes statistics, and is regarded in many hockey circles as on of the biggest up and comers in all of NA. Furthermore he is not expected to manage the maple leafs, he is expected to assist the GM, meaning that he will have final say in next to nothing.

Link to comment
Share on other sites

SO..... could simply have a Nerd Brilliant Biostatistician, crunching #s with a super computer in his mother's basement somewhere and get same thing.

Link to comment
Share on other sites

Basically nothing is the be all end all? Advanced stats won't tell you everything? Size won't tell you everything? CHaracter won't tell you everything? Skill won't tell you everything? The list could go on and on I guess. Sometimes too much emphasis is put on one aspect, when really a team (and a player) needs a balance of both. I can't yet tell if GMs have known this for years or if a new game is being played. All I know is they'd do a better job than me, so I'll try to keep my complaining to a minimum.

Link to comment
Share on other sites

Advanced Stats are universally related to possession of the puck.

The Habs proved last year that you can build a successful team built around capitalizing on the mistakes of your opponents, keeping them to the outside, preventing good scoring chances, and pouncing on it whenever they make a mistake. That type of strategy will, of necessity, have lousy possession stats, but it can be just as successful as brute-force controlling the play. As such, I feel that they can be useful, but some people are putting too much emphasis on them.

Link to comment
Share on other sites

I kept laughing at one unnamed (Eyes on the Prize) website that kept ranting Habs cant win, are only temporarily winning with smoke and mirrors, because this stat or that graph shows they suck (also common theme was; the coach and GM are idiots, are driving team into the ground and I hate the Habs because Douglas Murray is playing)

But surprise surprise, they actually had an OK season, when all the cherry picked advanced stats argued the opposite.

So now when I see post start rambling Corsi this or toss up some goofy graphs, I quickly lose interest knowing anyone can spin stats 1,000 different ways to fit your hypothesis or theory.

Link to comment
Share on other sites

Advanced Stats are universally related to possession of the puck.

The Habs proved last year that you can build a successful team built around capitalizing on the mistakes of your opponents, keeping them to the outside, preventing good scoring chances, and pouncing on it whenever they make a mistake. That type of strategy will, of necessity, have lousy possession stats, but it can be just as successful as brute-force controlling the play. As such, I feel that they can be useful, but some people are putting too much emphasis on them.

Those stats will go up this year with the additions of Gilbert and Beaulieu. Bouillon, Gorges and Murray couldn't move the puck.

Murray in particular was disastrous because of his QoC and his zone starts. He was given sheltered minutes and zone starts but somehow got pinned in his own end. I don't think he's as bad as some say, but there's a trickle down effect with a player like that, or a goon that the rest of the team suffers.

Link to comment
Share on other sites

Another thing to consider is that the advanced stats that some teams are using are proprietary and aren't anything that's out in the public domain. Who has a better chance at creating/identifying those stats, those who have studied new hockey stats (such as the recent blogger hires) or a general statistician? The hockey bloggers can better contextualize the sport-specific stuff compared to a recent college graduate simply with a stats degree. Hiring the latter doesn't even register for most people but the former gets more attention; it doesn't make it a PR move though.

Link to comment
Share on other sites

I'm sure these stats (which I can't be bothered with, personally) are one more useful piece of info, in a business in which people should be looking to minimize uncertainty to the greatest extent possible. That said, a free-flowing sport like hockey has many, many more intangibles than baseball, and I think the team that over-relies on these stats will rue the day. Useful, not the be-all and end-all.

Link to comment
Share on other sites

Another thing to consider is that the advanced stats that some teams are using are proprietary and aren't anything that's out in the public domain.

I read an article in SI awhile back on Dave Tippet, his stat du jour was passes before a shot. You're right about the stats being proprietary, but that's why I laugh about the "advanced" or "fancy stats" misnomer. People who think that they have access to this information that the GMs or coaches do not is laughable.

His QofC eh? Nevr considered that when he was running Lucic over and putting Chara on his ass!

That makes a big difference. We had to put Murray out against the fourth lines or when there was an offensive zone start. Murray was very entertaining and added a needed physical element, but that was no different from what we would get from a traditional enforcer, of which i'm always in favor. The games when Murray and Parros were playing were awful. That sequence before the Matt Fraser's overtime goal was an abomination.

QoC and zone starts are interesting stats. I like to see how a coach deploys players and with whom. With Or With Out You, or WOWY is another great stat. Most of the players have only a negligible difference, which I guess is that fault of all the "fancy stats" we see. They seem to be used exclusively to denigrate fourth liners. Just a bunch of wimps picking on the enforcers because they spent middle school in the nurse's office with bloody noses from dodgeball.

Link to comment
Share on other sites

I'm sure these stats (which I can't be bothered with, personally) are one more useful piece of info, in a business in which people should be looking to minimize uncertainty to the greatest extent possible. That said, a free-flowing sport like hockey has many, many more intangibles than baseball, and I think the team that over-relies on these stats will rue the day. Useful, not the be-all and end-all.

Sean McIndoe did a good article about how if you don't care about extra statistics, that's 100% fine. Nobody should be yelling at you to follow them. However, you can't say they don't have some effectiveness just because you don't wish to follow them.

A few years back, it was only Oilers bloggers and posters caring about advanced statistics and year after year, it did nothing for them to gauge the quality of their team. It was mostly to make excuses about Ales Hemsky's effectiveness. But soon every team was considering them and seeing new details.

The problem is that far too many people use the stats to win arguments. That's their only purpose out of them. More things to point at and argue their point. And really, those folks will jump on anything. It was no different in baseball. There are still Oakland A fans who believe bunting/stealing is worthless when the A's are now bunting/stealing a lot more than other teams because Beane said stats are saying it's good when used effectively. That's the key word. Effectively.

We're going to get to a point where all 30 teams are going to try playing a possession game and then a smart team that was already ahead of the curve like Los Angeles is going to completely change their game up specifically to break possession puck teams and they will easily win a Cup out of it. It's no different to what happened in the trap era. Everyone started trapping all game so certain teams like Tampa Bay and Anaheim learned to speed through it and go back into a trap after gaining a margin lead. The trap is still used but cycle and possession is making it only useful in defensive situations.

As much as I rag on a lot of the advanced stats guys and I truly think they are too obsessed over specific statistics, there is one thing they are doing that's going to make NHL discussion far better: they are destroying the +/- and the goalie wins basic stats. They have always been the most misleading, abused numbers in hockey and they need to go.

Link to comment
Share on other sites

Sean McIndoe did a good article about how if you don't care about extra statistics, that's 100% fine. Nobody should be yelling at you to follow them. However, you can't say they don't have some effectiveness just because you don't wish to follow them.

A few years back, it was only Oilers bloggers and posters caring about advanced statistics and year after year, it did nothing for them to gauge the quality of their team. It was mostly to make excuses about Ales Hemsky's effectiveness. But soon every team was considering them and seeing new details.

The problem is that far too many people use the stats to win arguments. That's their only purpose out of them. More things to point at and argue their point. And really, those folks will jump on anything. It was no different in baseball. There are still Oakland A fans who believe bunting/stealing is worthless when the A's are now bunting/stealing a lot more than other teams because Beane said stats are saying it's good when used effectively. That's the key word. Effectively.

We're going to get to a point where all 30 teams are going to try playing a possession game and then a smart team that was already ahead of the curve like Los Angeles is going to completely change their game up specifically to break possession puck teams and they will easily win a Cup out of it. It's no different to what happened in the trap era. Everyone started trapping all game so certain teams like Tampa Bay and Anaheim learned to speed through it and go back into a trap after gaining a margin lead. The trap is still used but cycle and possession is making it only useful in defensive situations.

As much as I rag on a lot of the advanced stats guys and I truly think they are too obsessed over specific statistics, there is one thing they are doing that's going to make NHL discussion far better: they are destroying the +/- and the goalie wins basic stats. They have always been the most misleading, abused numbers in hockey and they need to go.

Habs v Bruins provided the blueprint for that. Get in the lanes and block shots when they set up. A team without a good goalie won't succeed that way, but MTL did. The Bruins were really effective off the cycle when they had Horton and Iginla. I don't think their first line is going to be half as effective without that two headed monster down low.

Link to comment
Share on other sites

Actually if you look at the "close" stats in that series (ie game was tied or 1 goal lead), and take a look at the times when Murray was not on the ice... you can actually see that in the majority of games the Habs were the better puck possession team.

The idea that we only scored by capitalizing on Bruins mistakes and not puck possession is not true.

The Habs won that series by having a slight edge in possession when the game close... Having a better goalie... and when they got a two goal or more lead, thats the point the bruins started to control possession and the Habs started to sit back. This is the story in 3 of the Habs 4 wins...

It was a major reason I argued for Beaulieu after the Habs game 4 loss a game they dominated possession when Murray was on the bench (and were dominated by Boston when Murray was on the ice, leading to the goal against).

The Habs best players (Subban, Pacioretty, Plekanec, Markov) are huge drivers of puck possession.... the players we've gotten rid of this off-season (Parros, Murray, Bouillion, Briere, Gorges) are all negative influences on possession; and the players we've brought in (Parenteau, Gilbert, re-signing Weaver) are all possession drivers. Malholtra is not a possession driver (he's still a negative) but he's an upgrade on Ryan White.

Here is what we know.... Bergevin is a disciple of the Blackhawks... the Blackhawks are one of the NHL's leaders in utilizing advanced stats in the NHL.... all of Montreal's moves this off-season improved the club's advanced stats.... If you think that the Habs aren't aware of and using these stats, you are kidding yourself.

Link to comment
Share on other sites

Actually if you look at the "close" stats in that series (ie game was tied or 1 goal lead), and take a look at the times when Murray was not on the ice... you can actually see that in the majority of games the Habs were the better puck possession team.

The idea that we only scored by capitalizing on Bruins mistakes and not puck possession is not true.

The Habs won that series by having a slight edge in possession when the game close... Having a better goalie... and when they got a two goal or more lead, thats the point the bruins started to control possession and the Habs started to sit back. This is the story in 3 of the Habs 4 wins...

It was a major reason I argued for Beaulieu after the Habs game 4 loss a game they dominated possession when Murray was on the bench (and were dominated by Boston when Murray was on the ice, leading to the goal against).

The Habs best players (Subban, Pacioretty, Plekanec, Markov) are huge drivers of puck possession.... the players we've gotten rid of this off-season (Parros, Murray, Bouillion, Briere, Gorges) are all negative influences on possession; and the players we've brought in (Parenteau, Gilbert, re-signing Weaver) are all possession drivers. Malholtra is not a possession driver (he's still a negative) but he's an upgrade on Ryan White.

Here is what we know.... Bergevin is a disciple of the Blackhawks... the Blackhawks are one of the NHL's leaders in utilizing advanced stats in the NHL.... all of Montreal's moves this off-season improved the club's advanced stats.... If you think that the Habs aren't aware of and using these stats, you are kidding yourself.

The players coming in versus going out have more to do with being upgrades than pandering to Fenwick/Corsi.

If the Habs play keep away/the cycle/carry the puck into the zone instead of dump and chase, then I'll be convinced that there's more emphasis on fancy stats. That's not what we saw last year with regards to systems.

The logical fallacy is that the best teams are the best at Fenwick because they are the best teams. They aren't the best teams because they are the best at Fenwick.

Link to comment
Share on other sites

You gotta have talent behind your possession.

Everyone picked the Senators to destroy last season because they were young and had great corsi and qualcom stats. Instead they were mediocre at best. And it wasn't just "bad luck" like what tends to be recorded for possession teams. On paper, replacing Daniel Alfredsson and Jacob Silferberg with Bobby Ryan and Clark MacArthur sounds like the right thing to do. It ended up being disastrous.

Link to comment
Share on other sites

The players coming in versus going out have more to do with being upgrades than pandering to Fenwick/Corsi.

If the Habs play keep away/the cycle/carry the puck into the zone instead of dump and chase, then I'll be convinced that there's more emphasis on fancy stats. That's not what we saw last year with regards to systems.

The logical fallacy is that the best teams are the best at Fenwick because they are the best teams. They aren't the best teams because they are the best at Fenwick.

The Stanley Cup Champion has come out of a top 5 possession team in every season that these stats were recorded (going back to the early 2000s) except one. That one was the 2009 Pittsburgh Penguins... who were a top 5 possession team in the second half of the season (after gonchar returned, and after Therrien was fired).

Its not a fallacy to say that teams that possess the puck more than their opponents and generate more shot attempts are usually the best teams. The stats bear that out, and they also show that possession is a repeatable statistic over large sample sizes.

Yes, you need talent to generate good possession.... but the object of getting talent is also to improve possession. Its the chicken and the egg.

Upgrading your talent is likely to lead to better possession stats.....

Players with better possession stats are likely to be upgrades in talent (not just looking at one stat but the context of all stats combined and also scouting by eyeball).

Its not like I'm saying montreal is abandoning their scouting... no team is... but at the same time they are using these stats, its pretty clear knowing where Bergevin came from (Chicago) and the moves he's making now.

=================

As for the Habs system last year..... again they do much better in two instances.

1) When neither Bouillion or Murray were on the ice (fixed that problem).

2) When the score of the game was close (within two goals).

What this tells us is that yes, I expect the Habs to be a better possession team this year... especially with the roster that Bergevin is putting together.

Therrien has had good possession teams in his past... He's not a Carlyle where every team he has regresses possession wise year after year after year.

Therrien's 2008 Penguins were very good possession wise... His 2012-13 Habs were good in possession too, the 2013-14 team was good when the liabilities were removed. ... All of which makes me think that the 2009 result was more because he really missed gonchar who was still an elite defender at that time then anything else.

Link to comment
Share on other sites

You gotta have talent behind your possession.

Everyone picked the Senators to destroy last season because they were young and had great corsi and qualcom stats. Instead they were mediocre at best. And it wasn't just "bad luck" like what tends to be recorded for possession teams. On paper, replacing Daniel Alfredsson and Jacob Silferberg with Bobby Ryan and Clark MacArthur sounds like the right thing to do. It ended up being disastrous.

Actually some of the people who looked at fancy stats for the Sens predicted better possession but a worse record. Why? Because they knew that Craig Anderson wasn't repeating his .941 save percentage and Robin Lehner wasn't repeating his .936 save percentage from 2012-13. Those who predicted the great season made the mistake of just looking at Corsi/Fenwick.

Anything over 930 is pretty unsustainable unless you think you have one of the NHL's best goalies.

You can't do that.... you need to look at PDO (shooting and save percentage), Zone Starts, and if looking at a team special teams numbers as well.

In Montreal, I haven't worried too much about special teams because we always seem to be in the top half of the league in that given our current personnel.

Link to comment
Share on other sites

I have to say that Commandant has summed up the importance and relevance of the fancy stat in relation to the current Habs, in the best way for me to understand it, of anything I've ever read.... short and sweet, like my attention span :halm:

Link to comment
Share on other sites

I too enjoyed Commandant's analysis (as I usually do), and find it convincing. To my mind, it's a good thing that Zoot Suit is up to speed on this advanced stats stuff, because I hate the thought of the Habs been hidebound with "old school" mentalities. It's equally important not to go too far in the other direction, but a team that clings to old verities as times change is not going to go anywhere.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...