No, he believes it.

I ain't bashing him. And I've even come around on a few things . But the very fundamental side of this debate is probably something we'll never see eye to eye on
- HB77

Oh I get that he believes it because it is just his whole personality. I am like you there is a place for these stats and yes Gardiner is better than most people think. But to then go on and say the sh1t he does about Doughty just makes him come of as a clown.

Do you even know what hypocrisy is?

Let me quote my man Emerson for you "A foolish consistency is the hoggoblin of the mind....."

But I don't even think I'm being inconsistent. I specifically call for a non-extreme not black/white either/or approach.

If anyone wants to actually talk about the article I am happy to.
- James_Tanner

Of course I know what hypocrisy is. I read your articles.

The value of using standardized assessment tools is to refine the process and produce better than random results. That doesn't happen when the tools aren't applied consistently so it's not a "foolish consistency." It serves a logical purpose. I'm still waiting for the data that clearly supports your opinions that Hall>Duchene and RNH>Johansen. You know, since your opinions are always based on evidence like you said.

Except that if that makes you a better player then you will have better stats in things that correlate to winning. How do you know that Doughty's heart doesn't enable him to suppress two shots per game and that they are already showing up in the stats?

Intangibles are real, but they aren't bonuses added after all other things have been accounted for.
- James_Tanner

I don't disagree that stats (basic and advanced) are helpful to evaluate the skill and value of a player in certain situations. However, there is no stat or combination of stats that exist that can tell you that "player X is better than player Y". You can use Corsi to compare players at how good they are at generating Corsi events vs allowing Corsi events against, but that only gives a very small insight into the picture. Ignoring stats completely and relying on the "eye test" is a mistake, but to assess the quality of a player solely based on numbers on paper is just as big of a mistake.

For me this has been Gardiner's best year, whether its due to maturity or Babs influence or a bit of both he has improved his play based on the games I have watched. From an analytical aspect I have no idea where he stands comparitevely to his previous years, however if I were to compare him to a player he reminds me of a PK Subban lite with regards to his ability to skate - and sometimes over skate the puck similarly to how Subban would during his first few seasons. Now Jake is only a year younger than PK, but he has also spent less time in the NHL.

No, he believes it.

I ain't bashing him. And I've even come around on a few things . But the very fundamental side of this debate is probably something we'll never see eye to eye on
- HB77

Which is fine with me because you always at least listen to other sides of an argument and present your views respectfully. I am a huge fan of when you comment, even though we always disagree.

James Tanner: Gardiner, Doughty and the Case of the Advanced Stats Bandit
Advanced stats lead us to make some seemingly strange arguments.
- James_Tanner

Of course I know what hypocrisy is. I read your articles.

The value of using standardized assessment tools is to refine the process and produce better than random results. That doesn't happen when the tools aren't applied consistently so it's not a "foolish consistency." It serves a logical purpose. I'm still waiting for the data that clearly supports your opinions that Hall>Duchene and RNH>Johansen. You know, since your opinions are always based on evidence like you said.
- jfkst1

You can go to ownthepuck.blogspot and compare season by season between RNH and Johansen. RNH has had a much better career, though is not having as good of a season.

Tanner, you do understand that purely shot based metrics to quantify Defencemen's ability is not the most accurate representation on someones talent.

Currently thats the only data collection method, however I do not believe its paints the complete picture of a players ability, as it only takes Offence zone and Defensive zone shot situations into account.

Which is fine with me because you always at least listen to other sides of an argument and present your views respectfully. I am a huge fan of when you comment, even though we always disagree.
- James_Tanner

Stroke stroke

Good blog

Am looking forward to the Leafs trading Gardiner at the deadline for a 1st line center or a couple of first round picks.

I have a question, since we're talking science.

Where does all the data come from that populates all of these new found charts? I'm actually interested because i know in my professional life, when i use someone else's data to support my own hypotheses, I know damn sure who collected it, how, and that it's valid before I put my name on it.

I think that's the main claim against some of your theories, that you see some new stat based comparison and just accept that it's true without any kind of validation, or as it known scientifically, QA/QC (Quality Assurance/Quality Control)....even when it doesn't match up against what we see with our eyes, or the best judge of all, the actual outcome of games/season.

I don't disagree that stats (basic and advanced) are helpful to evaluate the skill and value of a player in certain situations. However, there is no stat or combination of stats that exist that can tell you that "player X is better than player Y". You can use Corsi to compare players at how good they are at generating Corsi events vs allowing Corsi events against, but that only gives a very small insight into the picture. Ignoring stats completely and relying on the "eye test" is a mistake, but to assess the quality of a player solely based on numbers on paper is just as big of a mistake.
- tkecanuck341

Agree 100%. Another thing that bothers me about these stats is they are not an apples to apples comparison. Every team doesn't play every other team an equal amount of time so that skews the numbers.

James Tanner: Gardiner, Doughty and the Case of the Advanced Stats Bandit
Advanced stats lead us to make some seemingly strange arguments.
- James_Tanner

you my friend are the very embodiment of a HockeyBuzz blogger

I have a question, since we're talking science.

Where does all the data come from that populates all of these new found charts? I'm actually interested because i know in my professional life, when i use someone else's data to support my own hypotheses, I know damn sure who collected it, how, and that it's valid before I put my name on it.

I think that's the main claim against some of your theories, that you see some new stat based comparison and just accept that it's true without any kind of validation, or as it known scientifically, QA/QC (Quality Assurance/Quality Control)....even when it doesn't match up against what we see with our eyes, or the best judge of all, the actual outcome of games/season.
- nikel

The stats come from the NHL. Even if they are not 100% accurate, they're good enough.

The Leafs of the past 3 seasons must be the unluckiest team in the history of hockey.

They have the equivalent of Drew Doughty on defense and the equivalent of Steven Stamkos as a 1st/2nd line center in Kadri, and yet somehow their won-loss record in this era is in the bottom 5 of the league. Bummer.

AEG selected by Glendale to run the Gila Monster Arena.

No details released.

The Leafs of the past 3 seasons must be the unluckiest team in the history of hockey.

They have the equivalent of Drew Doughty on defense and the equivalent of Steven Stamkos as a 1st/2nd line center in Kadri, and yet somehow their won-loss record in this era is in the bottom 5 of the league. Bummer.
- Snowblind

It's not bad luck.

Bettman fixes the games to make Toronto lose.

AEG selected by Glendale to run the Gila Monster Arena.

No details released.
- Atomic Wedgie

Good Coyotes blog. Best one in a long time.

You can go to ownthepuck.blogspot and compare season by season between RNH and Johansen. RNH has had a much better career, though is not having as good of a season.
- James_Tanner

I invite anyone to go through the Warrior and HERO charts on that site and conclude RNH is clearly "much better" than Johansen. Johansen has a better HERO chart which tracks the last three seasons and comparable Warrior charts over the same time.

Gardiner's a decent defenseman but the Warrior chart is misleading in that it might make people who don't understand the underlying measures think Gardiner is better.

Doughty puts up superior ice time, goals, assists and points.
Gardiner puts up superior CF60 RelTM, CA60RelTM and CD60 RelTM (all stats measured relative to teammates).

LA is currently ranked #1 in the league with a team 54.8% CF, a full 3.5% ahead of the Leafs. It's no surprise that Gardiner shows superior CF60 / CA60 / CD60 RelTm to Doughty because Gardiner plays on a team that has some really poopty players on it. Likewise, it's really tough for Doughty to put up great numbers because his team is so very, very good from a Corsi perspective.

If you compare them directly I think you get a better idea of how dominant each is - Doughty is rocking an ungodly 58.2 CF% at 5v5, good for fifth overall in the league and first amongst defensemen, minimum 500 mins played. (Coincidentally, LA players occupy the top 7 spots leaguewide in that statistic, and 8 of the top 10). Meanwhile, you have Gardiner sitting at a respectable 52.8 CF%, good for 108th overall and 35th amongst defensemen. All of a sudden he's not looking quite on par with Doughty...

The stats come from the NHL. Even if they are not 100% accurate, they're good enough.
- James_Tanner

You know the "enhanced stats" on NHL.com are ridiculed for being wildly inaccurate right?

I have a question, since we're talking science.

Where does all the data come from that populates all of these new found charts? I'm actually interested because i know in my professional life, when i use someone else's data to support my own hypotheses, I know damn sure who collected it, how, and that it's valid before I put my name on it.

I think that's the main claim against some of your theories, that you see some new stat based comparison and just accept that it's true without any kind of validation, or as it known scientifically, QA/QC (Quality Assurance/Quality Control)....even when it doesn't match up against what we see with our eyes, or the best judge of all, the actual outcome of games/season.
- nikel

Like each referee calls a game differently, even though they are suppose to abide by the same rules, many of these stats have a level of subjectivity by the person recording each event.

I have stated from the start, these stats are nothing more than directional, which should be used as a starting point of a discussion rather than the whole discussion.

There is value in these stats, but time and time again, Tanner has proven in his "analysis" that they are often not a true overall indication of the value/worth of a player.

Be a statshead all you want, but poke your head out of the numbers once and a while to catch a whiff of reality, too.

The stats come from the NHL. Even if they are not 100% accurate, they're good enough.
- James_Tanner

That is like saying every ref calls a game the same way, simply because they all use the same rule book.

There is a level of subjectivity / error in these numbers that is never accounted for... so no, they are not good enough to be the sole reasoning on the value of a player.

They are solely directional indicators that may lead to an understanding of the value of a player.

Like each referee calls a game differently, even though they are suppose to abide by the same rules, many of these stats have a level of subjectivity by the person recording each event.

I have stated from the start, these stats are nothing more than directional, which should be used as a starting point of a discussion rather than the whole discussion.

There is value in these stats, but time and time again, Tanner has proven in his "analysis" that they are often not a true overall indication of the value/worth of a player.

Be a statshead all you want, but poke your head out of the numbers once and a while to catch a whiff of reality, too.
- Mashadar

I am not saying Gardiner is better than Doughty.

But I am saying that if people who ignore stats have an opposite opinion of those who do use stats, then the stats people are always closer to being right.

Gardiner may not be Doughty, but the fact that you can make such a comparison does quantify him as an elite NHL dman. And this is necessary to point out because people try to tell me every day that he's a "decent #3 or 4"

That is like saying every ref calls a game the same way, simply because they all use the same rule book.

There is a level of subjectivity / error in these numbers that is never accounted for... so no, they are not good enough to be the sole reasoning on the value of a player.

They are solely directional indicators that may lead to an understanding of the value of a player.
- Mashadar

But they would never be far enough off reality to make a big difference in the aggregate.