NHL on brink of intelligence explosion

[schuckers]

Here's a link to an article in today's Boston Globe about a hockey panel last week at the Joint Statistical Meetings in Boston.

NHL on Brink of Intelligence Explosion

Full Disclosure: I was on the panel and I'm mentioned in the article.

Click to read more...

 

[Jacob8hockey*]

Interesting read, also interesting that MacDonald who was also on that panel just got hired by the Panthers.

 

[Mathletic]

Interesting read, also interesting that MacDonald who was also on that panel just got hired by the Panthers.

interesting, very good hire

...

analytics might not be only for the drunks after all

 

[TOML]

They understand that four lines of skill, speed, and puck-possessing prowess overwhelm the traditional template of two skilled units, a checking threesome, and an energy group. They acquire and play mobile, pace-pushing blue liners over hold-your-ground defensive defensemen. They don’t panic when they fall behind, 2-0, because statistics show that scoring next impacts the outcome more than allowing a third goal.

Duncan Keith is Chicago’s best defenseman. But Keith sits against first lines. Chicago deploys Niklas Hjalmarsson in a shutdown role. That allows Keith to flourish in situations that play to his strengths: quickness, speed with the puck, accurate passing, and offensive instinct.

To me this sounds like they value hold-your-ground defensive defensemen quite a bit.

Deploying a 'pace-pushing' guy against 1st liners is risk-reward and most of the time the risk outweighs the reward.


I'd agree that skill and speed are important for four lines, but not at the expense of defensive acumen. If LA and Chicago are the comparables, it should be noted that LA used a bunch of wingers who aren't particularly fast, but they aren't detriments without the puck, and Chicago's winning teams also had considerable size up front to match their skill and speed.

What's common though is they had great two-way depth up the middle. So not only did they convince their stars centermen to play great two-way hockey so they can be deployed in any situation, but they also complemented these players with more of the same type so ultimately all four lines could be deployed in any situation.

This would make the jobs of opposing coaches far more difficult, as there would be line-matching advantage to be gained, and their comparatively one-dimensional offensive stars would be exposed.

If teams just focused on speed, skill, possession, taking advantage of qualcomps with little regard to defensive acumen, 'grit', size etc., esp. up the middle, then they will find their teams to be great in the regular season, but not able to get over the hump in the playoffs.


There is no coincidence that the Hawks and Kings have been the best post-cap teams so far and they also recently have had remarkable player development to keep their lineups fresh, not to mention the young franchise players around to make their prime years sustainable longer.


Really, this approach to winning isn't new at all, as all great teams in the past have done this. I suppose analytics is a way to verify that what worked before still works today. More importantly, it quantifies it, so teams/coaches aren't blind to what players will work and what won't. Really, it should eliminate any excuse for bad decision-making. Like bad signings. Yet, I fully expect bad decisions to be made all the time. After all, there's only one winner per season.

 

[Viktri]

As a huge stats guy , I can't say that I agree with the article.

Stats help evaluate strategies - this is true. Strategies being born from stats - this is unlikely (it happens, but very uncommon). Stats being necessary to evaluate the effectiveness of strategies - absolutely untrue. Stats definitely help evaluate the effectiveness of strategies (and may overturn erroneous evaluations) but is not 100% required to do so.

Some of the examples cited were poorly though out. In what scenario is it appropriate for a coach to throw in the towel and concede defeat? After 1 goal, 2 goal, 3 goals, etc? In no scenario should the coach throw in the towel and in no scenario should they panic - stats don't help there.

Scoreless after 2 means that you should score in the 3rd - well, duh.

1-0 lead isn't safe; yes, especially considering the average goals per game is higher than this - but I didn't need additional new statistics to tell me this. Any lucky play could erase a 1 goal lead.

Personally, I think the revolution in the stats industry has more to do with how collecting/interpreting data is changing to become more granular to measure things on a team-specific basis; each team has different weaknesses and I'd wager that the new stat guys are being brought on board to 1) evaluate the weaknesses of the team (and quantify the magnitude of the weaknesses impact); 2) figure out how to measure which players have the skills/traits missing from the hockey club; 3) evaluate the magnitude of the skills/traits that those players have and 4) determine whether those players can fit into the team.

It's not something ground breaking but just an extension of what has been done before. Teams have always looked for players to fit into the team - they're just using additional tools to determine the best fit.

 

[Freudian]

SportVU is going to revolutionize stat tracking in the NHL. I suspect most of it will be kept internal by the teams and depending on how public the data will be, it might be hard to make it into something that enhances the fan experience.

 

[Beatle17]

To me this sounds like they value hold-your-ground defensive defensemen quite a bit.

Deploying a 'pace-pushing' guy against 1st liners is risk-reward and most of the time the risk outweighs the reward.


I'd agree that skill and speed are important for four lines, but not at the expense of defensive acumen. If LA and Chicago are the comparables, it should be noted that LA used a bunch of wingers who aren't particularly fast, but they aren't detriments without the puck, and Chicago's winning teams also had considerable size up front to match their skill and speed.

What's common though is they had great two-way depth up the middle. So not only did they convince their stars centermen to play great two-way hockey so they can be deployed in any situation, but they also complemented these players with more of the same type so ultimately all four lines could be deployed in any situation.

This would make the jobs of opposing coaches far more difficult, as there would be line-matching advantage to be gained, and their comparatively one-dimensional offensive stars would be exposed.

If teams just focused on speed, skill, possession, taking advantage of qualcomps with little regard to defensive acumen, 'grit', size etc., esp. up the middle, then they will find their teams to be great in the regular season, but not able to get over the hump in the playoffs.


There is no coincidence that the Hawks and Kings have been the best post-cap teams so far and they also recently have had remarkable player development to keep their lineups fresh, not to mention the young franchise players around to make their prime years sustainable longer.


Really, this approach to winning isn't new at all, as all great teams in the past have done this. I suppose analytics is a way to verify that what worked before still works today. More importantly, it quantifies it, so teams/coaches aren't blind to what players will work and what won't. Really, it should eliminate any excuse for bad decision-making. Like bad signings. Yet, I fully expect bad decisions to be made all the time. After all, there's only one winner per season.

This whole post is the best I have read in regards to stats articles on most boards. The bolded part is excellent but will get angry responses from Stats Guys.

 

[NikF]

Willie Mitchell, Rob Scuderi, Robyn Regehr, Matt Greene are just some of the defensive defensemen the Kings Cup winning teams have used. In fact Dean Lombardi is on record as saying he prefers pairings where astay-at-homer is paired up with a puck-mover. So the narrative that they're moving away from that is a bit wrong. Nor is speed and skill exactly the term I'd use to describe the Kings, especially not their 2012 team.

 

[TOML]

Willie Mitchell, Rob Scuderi, Robyn Regehr, Matt Greene are just some of the defensive defensemen the Kings Cup winning teams have used. In fact Dean Lombardi is on record as saying he prefers pairings where astay-at-homer is paired up with a puck-mover. So the narrative that they're moving away from that is a bit wrong. Nor is speed and skill exactly the term I'd use to describe the Kings, especially not their 2012 team.

This. All these defense-first defenders, quite a few sluggish, grinding wingers, and they're still a badass possession team. But the buy-in from all their players like Kopitar, who wasn't always Selke-like, is vital. Winning teams have been doing this for decades.

 

[Brainiac]

Well, of course, with Don Cherry's incoming retirement, you can expect an "intelligence" explosion!

Joking aside, I've been looking forward to better possession stats. Will be interesting.

For example, with that kind of data, we'll likely be able to improve our analysis of giveaways and takeaways (which are, as of now, subjectively measured anyways).

For example, Karlsson takes the puck in his own zone, rushes it up the ice and then gives it away in the opponent's zone. For comparison, let's say a plug 4th liner gives the puck away in neutral zone, which then results in an instant scoring chance against.

Both plays are recorded as giveaways on the stat sheet. But one is starting from his own zone, ends up the offensive zone and might create a scoring chance in the process. The other starts in the neutral zone, finishes in the defensive zone and results in a scoring chance against.

With better possession data, we'll be able to actually measure the positive impact of puck carriers and discriminate them much better from the real turnovers machines.

 

[badtakemachine]

I'm still not entirely convinced that possession = winning. Yes, of course I've seen the plots tracking Corsi and Fenwick compared to win percentage, but those plots certainly have their outliers. As an Avs fan, the skepticism surrounding the "possession is everything" mantra is of course heightened. I think the Avs are a great example that possession is not everything. I think a guy named Dave Johnson did some preliminary work tracking "shots off the rush", i.e. in scoresheet-scraping terms, shot attempts within ten seconds of an event in the neutral zone or opposite attack zone. I think this type of work is equally important to build the foundations for an "intelligence explosion". Attempting to judge a transition team using a possession based scoring system is prone to give bad predictions and seems to be a square-peg-circular-hole type of problem. In hockey it wouldn't be this extreme, but it has the same feel as trying to quantify the strength of the Nash/Dantoni 7-seconds-or-less Suns using a half court, run the shotclock model.

 

[Bank Shot]

I'm still not entirely convinced that possession = winning. Yes, of course I've seen the plots tracking Corsi and Fenwick compared to win percentage, but those plots certainly have their outliers. As an Avs fan, the skepticism surrounding the "possession is everything" mantra is of course heightened. I think the Avs are a great example that possession is not everything. I think a guy named Dave Johnson did some preliminary work tracking "shots off the rush", i.e. in scoresheet-scraping terms, shot attempts within ten seconds of an event in the neutral zone or opposite attack zone. I think this type of work is equally important to build the foundations for an "intelligence explosion". Attempting to judge a transition team using a possession based scoring system is prone to give bad predictions and seems to be a square-peg-circular-hole type of problem. In hockey it wouldn't be this extreme, but it has the same feel as trying to quantify the strength of the Nash/Dantoni 7-seconds-or-less Suns using a half court, run the shotclock model.

You have to admit that everything went right for the Avs last season.

They outperformed their goal differential. They had the 8th best goal diff but finished 3rd in points.

They had the best record in the league in OT/shootout games, winning 15.

1st in the league in winning percentage in one goal games.

5th best PP.


Their goalie smashed his personal bests stats wise.


Pretty much all of those things seem to have a fair bit on randomness in them. It was just an extraordinary amount of things that just worked out perfectly for the Avs.



I'm not saying possession is the be all and end all but it seems to have a lot more repeat-ability season over season then the other stats I posted.

If the Avs for instance take a hit on goaltending back into the .915 range next season, their PP drops off, and their possession doesn't improve they will be in a world of hurt.

 

[badtakemachine]

You have to admit that everything went right for the Avs last season.

They outperformed their goal differential. They had the 8th best goal diff but finished 3rd in points.

They had the best record in the league in OT/shootout games, winning 15.

1st in the league in winning percentage in one goal games.

5th best PP.


Their goalie smashed his personal bests stats wise.


Pretty much all of those things seem to have a fair bit on randomness in them. It was just an extraordinary amount of things that just worked out perfectly for the Avs.



I'm not saying possession is the be all and end all but it seems to have a lot more repeat-ability season over season then the other stats I posted.

If the Avs for instance take a hit on goaltending back into the .915 range next season, their PP drops off, and their possession doesn't improve they will be in a world of hurt.

My post was not to try to prove the Avs didn't overachieve last season, because I believe they did too. However, I don't buy that the Avs will 'regress' to the point of being a bottom ten team in the league because of their possession numbers. They haven't been built as a possession team, so they shouldn't be treated as one.

Regarding the stats you listed, yes, there are randomness in any stat. Do you believe that everybody on the roster, including MacKinnon, Duchene, O'Reilly, etc., will fail to hit the 30 goal mark again? Do you believe none of these players will surpass 70 points again? I don't believe the PP is due to 'regress' either - I think this one is a pretty controllable stat. I also don't think the timing of Semyon Varlamov's meteoric improvement was too big of a surprise: 12/13 - Joe Sacco "system", half-ass goalie coach from Kirk McLean --> 13/14 - Patrick Roy's system, full-time goalie coach from Francois Allaire, a big improvement over McLean. Maybe Varlamov won't be in the top three Vezina voting every season, but he won't have to be.

 

[Bank Shot]

My post was not to try to prove the Avs didn't overachieve last season, because I believe they did too. However, I don't buy that the Avs will 'regress' to the point of being a bottom ten team in the league because of their possession numbers. They haven't been built as a possession team, so they shouldn't be treated as one.

Regarding the stats you listed, yes, there are randomness in any stat. Do you believe that everybody on the roster, including MacKinnon, Duchene, O'Reilly, etc., will fail to hit the 30 goal mark again? Do you believe none of these players will surpass 70 points again? I don't believe the PP is due to 'regress' either - I think this one is a pretty controllable stat. I also don't think the timing of Semyon Varlamov's meteoric improvement was too big of a surprise: 12/13 - Joe Sacco "system", half-ass goalie coach from Kirk McLean --> 13/14 - Patrick Roy's system, full-time goalie coach from Francois Allaire, a big improvement over McLean. Maybe Varlamov won't be in the top three Vezina voting every season, but he won't have to be.

I just can't find any examples of teams that are bottom ten in possession metrics over multiple seasons and make the playoffs consistently. Maybe you can?

I'm not going to believe that the Avs are unique, and if they are bottom 5 in possesion metrics again next year it wouldn't surprise me one bit to see them miss the playoffs. They can always improve their possession stats of course.

I'm not sure PP% is a controllable stat. Pittsburgh has the two best players in the game, and while they are usually are a strong PP team, they had like the 19th best PP the year they won the cup, and they have dipped down to 20th or worse other seasons as well.

The Avs shooting percentage overall was 2nd in the league last season. Some teams are able to stay at the top for 2-3 seasons in a row but they will bounce from like 2nd to 5th or 6th. The difference from 2nd to 5th can mean a -10 goal difference.

If Varlamov's sv% drops to say .920 which is still very good. Like top ten starter good. If that happens the Avs get scored on 17 more times.

If those two very slight slips happen to the Av's percentages they go from +30 to +3 which is a borderline playoff team.

It doesn't take much at all to slip when a team is relying on having nearly the best shooting % and the best save % to win games.


I believe if the Avs are still a strong team next season it is because their possession stats have dramatically improved.

 

[Aceboogie]

Well, of course, with Don Cherry's incoming retirement, you can expect an "intelligence" explosion!

Joking aside, I've been looking forward to better possession stats. Will be interesting.

For example, with that kind of data, we'll likely be able to improve our analysis of giveaways and takeaways (which are, as of now, subjectively measured anyways).

For example, Karlsson takes the puck in his own zone, rushes it up the ice and then gives it away in the opponent's zone. For comparison, let's say a plug 4th liner gives the puck away in neutral zone, which then results in an instant scoring chance against.

Both plays are recorded as giveaways on the stat sheet. But one is starting from his own zone, ends up the offensive zone and might create a scoring chance in the process. The other starts in the neutral zone, finishes in the defensive zone and results in a scoring chance against.

With better possession data, we'll be able to actually measure the positive impact of puck carriers and discriminate them much better from the real turnovers machines.

I too would be interested in this sort of stat. Maybe a combination of zone entry and zone exit stats. If you have zone exit w/o a subsquent zone entry (either by you or teammate) shows you rushed the puck out (controlled exit), but lost the puck somewhere in the neutral zone. On the other hand if you have a number of giveaways but no controlled zone exits, it signals you are receiving the puck in the neutral zone, but losing it

 

[wgknestrick]

I'm still not entirely convinced that possession = winning. Yes, of course I've seen the plots tracking Corsi and Fenwick compared to win percentage, but those plots certainly have their outliers. As an Avs fan, the skepticism surrounding the "possession is everything" mantra is of course heightened. I think the Avs are a great example that possession is not everything. I think a guy named Dave Johnson did some preliminary work tracking "shots off the rush", i.e. in scoresheet-scraping terms, shot attempts within ten seconds of an event in the neutral zone or opposite attack zone. I think this type of work is equally important to build the foundations for an "intelligence explosion". Attempting to judge a transition team using a possession based scoring system is prone to give bad predictions and seems to be a square-peg-circular-hole type of problem. In hockey it wouldn't be this extreme, but it has the same feel as trying to quantify the strength of the Nash/Dantoni 7-seconds-or-less Suns using a half court, run the shotclock model.

Possession isn't everything, but you can say that about any factor because nothing is absolute. I like to think of it like this: There are multiple factors that contribute into the "100%" of winning a game.

• Possession has generally been proven (and accepted) as being the largest slice of that 100% pie chart. It is also accepted that this is the factor that is easier to predict.
  
• PDO%s (SH% and SV%) are generally the 2nd largest slice, but have a higher % variation

It is foolish to choose 1 vs the other IMO. They are BOTH important as proven by basic math.

X*Y=Z

X is no more important than Y in maximizing "Z". I think we all feel there are strategies to not only improve your # of shots (X), but also the shooting % (Y) of those shots to maximize goals. Teams just feel that it is easier to control (and track) the # of shots variable in that equation.

 

[badtakemachine]

Possession isn't everything, but you can say that about any factor because nothing is absolute. I like to think of it like this: There are multiple factors that contribute into the "100%" of winning a game.

Yeah, so exactly what I said. Possession is not everything.

 

[BB1133]

I'm still not entirely convinced that possession = winning. Yes, of course I've seen the plots tracking Corsi and Fenwick compared to win percentage, but those plots certainly have their outliers. As an Avs fan, the skepticism surrounding the "possession is everything" mantra is of course heightened. I think the Avs are a great example that possession is not everything. I think a guy named Dave Johnson did some preliminary work tracking "shots off the rush", i.e. in scoresheet-scraping terms, shot attempts within ten seconds of an event in the neutral zone or opposite attack zone. I think this type of work is equally important to build the foundations for an "intelligence explosion". Attempting to judge a transition team using a possession based scoring system is prone to give bad predictions and seems to be a square-peg-circular-hole type of problem. In hockey it wouldn't be this extreme, but it has the same feel as trying to quantify the strength of the Nash/Dantoni 7-seconds-or-less Suns using a half court, run the shotclock model.

New stats will be developed, but since Fenwick was tracked (2007), I believe no cup winner has been a sub 50% possession team, and it's highly correlated with making the playoffs over that time too.

I'd agree that I don't think the Avs are as bad as the fancy stats suggest, but they definitely aren't some kind of corsi-buster outlier either. The same scenario played out the last 2 years with the Leafs and we all know how that turned out.

A regression to the middle of the pack is almost inevitable...I just hope it's not worse than that.

 

[Rants Mulliniks]

New stats will be developed, but since Fenwick was tracked (2007), I believe no cup winner has been a sub 50% possession team, and it's highly correlated with making the playoffs over that time too.

I'd agree that I don't think the Avs are as bad as the fancy stats suggest, but they definitely aren't some kind of corsi-buster outlier either. The same scenario played out the last 2 years with the Leafs and we all know how that turned out.

A regression to the middle of the pack is almost inevitable...I just hope it's not worse than that.

They seemed to be a fairly good example that if you have other components that are strong, things like possession aren't so relevant but if those things dry up, possession can help you stay afloat.

 

[badtakemachine]

New stats will be developed, but since Fenwick was tracked (2007), I believe no cup winner has been a sub 50% possession team, and it's highly correlated with making the playoffs over that time too.

Yes, of course there is correlation between shot attempts and winning, but I don't think it is particularly surprising that those seven teams tended to outshoot their opponents. I bet you could say the same about simply counting goals for percentage, or even simply shots on goal percentage. If all seven of those teams were within the top five Fenwick teams each season, I would say that is much more compelling evidence.

I'd agree that I don't think the Avs are as bad as the fancy stats suggest, but they definitely aren't some kind of corsi-buster outlier either. The same scenario played out the last 2 years with the Leafs and we all know how that turned out.

A regression to the middle of the pack is almost inevitable...I just hope it's not worse than that.

While I do believe that the Avs overachieved in terms of points in the standings last season, I don't believe they are due for a big regression back to being a .500 team.

First, the whole point of Corsi and Fenwick is to create a situation where we have statistics that graduate from the 'small sample size' classification quicker than just counting goals or shots. After a full season of 82 games, don't goals for have just as much importance than the Corsi?

Second, why is it that a team can overachieve for an 82 game period, but cannot repeat that in game 83 (or game 90 if you count the playoff games)? Why does it have to reset in September? In the Leafs case, they clearly showed some drop off mid way through their season. In the Avs case, sure, they started at a ridiculous 13-1 or whatever, but they really had no bad stretches all season long, including their late season surge to take over the division title. This doesn't look like a team that is set up for a regression due to one large outlier stretch in the season.

Lastly, why is it that the only way for the supposed regression to take place is for the team's point total ranking to drop to its Corsi ranking? Why can't the team's Corsi ranking regress back up to match the team's point total? For the Avs this season, I believe it will be somewhere in the middle. After a year of development, I can hardly see this young core getting significantly worse.

 

[Freudian]

New stats will be developed, but since Fenwick was tracked (2007), I believe no cup winner has been a sub 50% possession team, and it's highly correlated with making the playoffs over that time too.

I'd agree that I don't think the Avs are as bad as the fancy stats suggest, but they definitely aren't some kind of corsi-buster outlier either. The same scenario played out the last 2 years with the Leafs and we all know how that turned out.

A regression to the middle of the pack is almost inevitable...I just hope it's not worse than that.

Avs corsi has been as wildly fluctuating as their results lately. They were 10th in ES corsi two years ago while finishing 20th in the league. They were 19th in ES corsi the year after when finishing 29th in the league and last season they were 25th in ES corsi while finishing third in the league.

Everyone who follows the team know that last years team was better than the team the previous two years and using corsi to evaluate them leads you to the wrong conclusions.

Of course Avs won't finish with 112 points again next year. I don't need stats to predict that. I'm using the old look-at-the-roster method combined with common sense.

 

[BB1133]

Yes, of course there is correlation between shot attempts and winning, but I don't think it is particularly surprising that those seven teams tended to outshoot their opponents. I bet you could say the same about simply counting goals for percentage, or even simply shots on goal percentage. If all seven of those teams were within the top five Fenwick teams each season, I would say that is much more compelling evidence.

2008 - DET 1st (59.6%)
2009 - PIT 14th (49.9% .. but were 54.9% after Bylsma took over for Therrien)
2010 - CHI 1st (57.8%)
2011 - BOS 14th (50.6%)
2012 - LA 4th (53.6%)
2013 - CHI 2nd (56.1%)
2014 - LA 1st (56.7%)

Seems compelling to me. Possession numbers really started to gain more clout once LA won the cup as an 8th seed when all the stat guys were saying they were they team to watch out for.

and they look at more than just FC% to predict team success. When you combine it with their PDO, it becomes really clear how a team is winning or losing. Together the predictive value is quite high.

Second, why is it that a team can overachieve for an 82 game period, but cannot repeat that in game 83 (or game 90 if you count the playoff games)? Why does it have to reset in September? In the Leafs case, they clearly showed some drop off mid way through their season. In the Avs case, sure, they started at a ridiculous 13-1 or whatever, but they really had no bad stretches all season long, including their late season surge to take over the division title. This doesn't look like a team that is set up for a regression due to one large outlier stretch in the season.

2nd in team shooting % 5on5, 2nd in team save % 5on5 + 27th in FC%. Success isn't sustainable with numbers like that. It's possible they remain a high shooting % team due to their skill up front, but even good teams fall off. Avs won a ton of 1 goal games last year, too (28-4-8). While it's possible Varly remains an elite goal, he's still due to fall back. They also won't start the season 13-1 or be catching teams off guard with their pulling-the-goalie-early routine.

Lastly, why is it that the only way for the supposed regression to take place is for the team's point total ranking to drop to its Corsi ranking? Why can't the team's Corsi ranking regress back up to match the team's point total? For the Avs this season, I believe it will be somewhere in the middle. After a year of development, I can hardly see this young core getting significantly worse.

This is where I disagree with the stats guys too. I think the team will continue to get better, especially if MacK breaks out. That will help their possession numbers and it's possible to improve towards that magical 50% number. I think we'll see both possession and point totals regress towards the middle next year and they wind up fighting for a wild card spot. But if they don't improve and remain a bottom 10 FC% team, there's a really high probability they miss the playoffs.

Avs corsi has been as wildly fluctuating as their results lately. They were 10th in ES corsi two years ago while finishing 20th in the league. They were 19th in ES corsi the year after when finishing 29th in the league and last season they were 25th in ES corsi while finishing third in the league.

Everyone who follows the team know that last years team was better than the team the previous two years and using corsi to evaluate them leads you to the wrong conclusions.

Of course Avs won't finish with 112 points again next year. I don't need stats to predict that. I'm using the old look-at-the-roster method combined with common sense.

Yea whether it's FC% or corsi, it doesn't really matter which metric is used, since they missed the playoffs in both those seasons with sub 50% possession rankings. That's not wild fluctuation. Last season was the 2nd time they performed like an outlier team in recent years.

Here are the Avs numbers since Fenwick has been tracked ...

2008 FC% 13th (50.3%) PDO 17th (1000) Standings 10th - made playoffs
2009 FC% 26th (46.4%) PDO 30th (976) Standings 28th - missed
2010 FC% 26th (46.9%) PDO 3rd (1016) Standings 12th - made playoffs
2011 FC% 23rd (47.6%) PDO 28th (984) Standings 29th - missed
2012 FC% 14th (49.8%) PDO 23rd (994) Standings 20th - missed
2013 FC% 26th (45.1%) PDO 28th (976) Standings 29th - missed
2014 FC% 27th (46.8%) PDO 3rd (1022) Standings 3rd - made playoffs


In 2008 they were a legit, but average, playoff team. They beat the overachieving Wild (FC% 23rd 48.1%, 5th in PDO) in the first round. Next round they got swept.

The last 2 times they made the playoffs (2010, 2014) they did it mostly with a really high PDO which masked their problems, and went out in the first round both times.

 

[badtakemachine]

Quickly looking at your data Brooks, it would appear that PDO correlates closer than Fenwick does. Has anyone looked at the R^2 comparing PDO to win percentage ranking vs Fenwick to win percentage ranking? I performed this analysis for GF% vs Fenwick vs Corsi and the R^2 values decreased in the order mentioned. Curious to see where PDO fits in. Perhaps the key to winning is actually get a couple of snipers and Tuukka Rask! Of course, that would also raise the question: how sustainable is PDO? Again, looking at the Avs only, neither Fenwick nor PDO appear to be very repeatable, but I'd be interested to see how teams that don't have such volatile roster turnover like COL recently fared season to season.

 

[Rants Mulliniks]

2008 - DET 1st (59.6%)
2009 - PIT 14th (49.9% .. but were 54.9% after Bylsma took over for Therrien)
2010 - CHI 1st (57.8%)
2011 - BOS 14th (50.6%)
2012 - LA 4th (53.6%)
2013 - CHI 2nd (56.1%)
2014 - LA 1st (56.7%)

Seems compelling to me. Possession numbers really started to gain more clout once LA won the cup as an 8th seed when all the stat guys were saying they were they team to beat.

You could just as easily look at goals against. In those same years the champs finished:

1, 17, 6, 2, 2, 1, 1

 

[Rants Mulliniks]

Quickly looking at your data Brooks, it would appear that PDO correlates closer than Fenwick does. Has anyone looked at the R^2 comparing PDO to win percentage ranking vs Fenwick to win percentage ranking? I performed this analysis for GF% vs Fenwick vs Corsi and the R^2 values decreased in the order mentioned. Curious to see where PDO fits in. Perhaps the key to winning is actually get a couple of snipers and Tuukka Rask! Of course, that would also raise the question: how sustainable is PDO? Again, looking at the Avs only, neither Fenwick nor PDO appear to be very repeatable, but I'd be interested to see how teams that don't have such volatile roster turnover like COL recently fared season to season.

Similarly, on GF only LA bucked the trend:

3, 6, 3, 5, 29, 2, 26

Worth noting too that LA didn't really "buck the trend" as in the playoffs they were 1st and 3rd in GF.

Long story short, teams that score a lot while giving up few tend to win, which shouldn't be surprising given the whole object is to outscore ones opponent.