• Recently Browsing   0 members

    No registered users viewing this page.

Richdg

Fixing this mess....

Rate this topic

Recommended Posts

I respect these advanced stats but my problem with these charts that pop up is they seem badly annotated to me. ...or I am an idiot. I'm open to that possibillity.

It says "deviation from NHL avg by line". and the colours show which line their getting compared to.

WIth that, my reading of those charts are:

- Glendening is being compared to the average 2nd liner in the 2nd chart? No wonder he's lower than the average. Especially since the chart shows he's being giving "buried" zone starts, which I assume means defensive starts. Of course a guy getting a majority of defensive zone starts will have a lower corsi than the average second line player who probably doesn't. Also, the 2nd chart says "all situations," so they're comparing his corsi including penalty killing time to the average 2nd liner, who probably doesn't PK and might be on the PP.

- Then, Tatar is being compared to 4th liners in the 1st chart? If so, no wonder he looks good. The biggest thing that tells me is we have to play Tatar more 5v5 than the average 4th liner.

I know the line classification is probably due to how much ice time they get, but it should be kept in mind.

The chart has buried and Sheltered as the labels for zone starts. I assume Sheltered means they get offensive starts, but getting defensive zone shouldn't be considered "buried." Glendening was one of our only guys could win faceoffs so he got the D zone starts. Since he apparently got 2nd line minutes, that hardly sounds buried to me.

If those are the perameters of the chart, I would question the conclusions of whether a player is struggling vs. thriving.

If I read it wrong, apologies, and feel free to correct me.

Edited by PavelValerievichDatsyuk

Share this post


Link to post
Share on other sites

PVD you have to take it into context.

You can't compare Tatar to Glendening in the "all situations" chart. You would have to compare Tatar to Nyquist, Datsyuk, Zetterberg and all the other players who get similar time of 5 on 5 time. In that regard, Tatar is obviously the best possession player.

Glendening's best comparison is probably someone like Darren Helm who gets PK time as well. In that regard Helm is the superior possession player.

As far as the all situations numbers are concerned, take it this way: Glendening is being put out in the defensive zone more often but is struggling. There are players in the league who have similar if not worse defensive zone starts than him who aren't struggling nearly as much and are on worse teams than the Red Wings (I've already posted this chart in the other thread). But to me this isn't about Glendening, it's more about the entire team. If it's a 3-2 game with 1 minute left to go in the favor of Red Wings, this chart makes me think that Glendening should not be on ice to take the faceoff as he struggles in the defensive zone. I'd much rather put another center to take the faceoff even with a slightly lower faceoff win %. Because what really is the difference between 54% and 53% faceoff and first guy being Glendening and the second player being Datsyuk?

Finally, there actually is no correlation to Glendening's zone starts and his possession numbers. In 2013 when he entered the league, he had about 50% defensive zone starts compared to last year where he had 70% defensive zone starts but his Corsi was only SLIGHTLY better in 2013 (but he played far less games in 2013 so it probably would have ended up being the same). It was still far far below the team average. So the issue with Glendening is not because of his zone starts, it's because of his inherent lack of possession skills. He's simply not a good possession player, regardless of what situation he's in.

Edited by kickazz

Share this post


Link to post
Share on other sites

Helm's big advantage on the defensive zone is his speed allows him to get the puck out once he has it on his stick. Glendening can't do that. Of course, once Helm does get the puck out of the zone, he has no clue or ability to do much offensively, which is what makes his contract so bad.

Share this post


Link to post
Share on other sites

I wanted to go back to a point someone made in another thread about how people who use advanced analytics are acting like they know better than the coaches. This is not true at all. There's things in Blashill's coaching which makes sense and actually works and nobody is refuting the point. The perfect example is Brenden Smith who was sheltered and thrived under Blashill. Another example is demoting Ericsson to third line and promoting Dekeyser to first line. I'd say most of the stuff put out on ice is just fine and obviously Blash and the analytics are lining up in that regard.

The issue comes in where certain MINORITY of the players are being misused and that's where the criticism starts to fall on Blash. Right now there's a few players being misused and those include Glendening, Zetterberg and Drew Miller.

So the next time someone comes about with wacky arguments that "Oh look at the analytics people trying to act like they know better". Stop seeing s*** in black and white, that's not what people are trying to say. They are just pointing out some of the weakest links in the lineup that need to be changed in order to make the team go from being an 8 seed to something better.

"Your team is only as strong as your weakest player"

Edited by kickazz

Share this post


Link to post
Share on other sites

PVD you have to take it into context.

You can't compare Tatar to Glendening in the "all situations" chart. You would have to compare Tatar to Nyquist, Datsyuk, Zetterberg and all the other players who get similar time of 5 on 5 time. In that regard, Tatar is obviously the best possession player.

Glendening's best comparison is probably someone like Darren Helm who gets PK time as well. In that regard Helm is the superior possession player.

I was trying to put it into context. My thing is "avg. by line" part of the chart. Looking at the first chart. They have Tatar in green which means he had the minutes of a 4th liner. I also think it means they are comparing his corsi% to how much it deviates from average corsi % of 4th liners not players like Nyquist, Z, DAts. If that's the case, that my issue with this chart. They could have done just corsi % as the bottom axis, and that would be more useful.

Also, on context with Glendening, you can see in the first chart that he's within the average corsi % despite the fact that he's shown to have the hardest (or buried) zone starts. That reflects well on him. In the second chart, where it shows all situations, he's well below the average, but "all situations" for Glen means it adds his PK time. PK obviously isn't going to be good for corsi rating because you're going to be stuck in your own zone and almost all the shot attempts will be at your net. You're right, you should compare him to Helm in this context. It's impressive that Helm (and Sheahan) are still within the average despite PK time, but I would expect PKers be below the average here.

P.S. Why isn't Miller on the chart? I know he only played around 30 games, but they have Ferraro and Pulkkinen only played 30-40. Miller would be an important comparison with Glendening.

Share this post


Link to post
Share on other sites

I respect these advanced stats but my problem with these charts that pop up is they seem badly annotated to me. ...or I am an idiot. I'm open to that possibillity.

It says "deviation from NHL avg by line". and the colours show which line their getting compared to.

WIth that, my reading of those charts are:

- Glendening is being compared to the average 2nd liner in the 2nd chart? No wonder he's lower than the average. Especially since the chart shows he's being giving "buried" zone starts, which I assume means defensive starts. Of course a guy getting a majority of defensive zone starts will have a lower corsi than the average second line player who probably doesn't. Also, the 2nd chart says "all situations," so they're comparing his corsi including penalty killing time to the average 2nd liner, who probably doesn't PK and might be on the PP.

- Then, Tatar is being compared to 4th liners in the 1st chart? If so, no wonder he looks good. The biggest thing that tells me is we have to play Tatar more 5v5 than the average 4th liner.

I know the line classification is probably due to how much ice time they get, but it should be kept in mind.

The chart has buried and Sheltered as the labels for zone starts. I assume Sheltered means they get offensive starts, but getting defensive zone shouldn't be considered "buried." Glendening was one of our only guys could win faceoffs so he got the D zone starts. Since he apparently got 2nd line minutes, that hardly sounds buried to me.

If those are the perameters of the chart, I would question the conclusions of whether a player is struggling vs. thriving.

If I read it wrong, apologies, and feel free to correct me.

Well you did read it wrong, but you're right that it is badly annotated.

Glendening and Tatar and everybody else's icetime is being compared to NHL line averages to determine whether their TOI is similar to the average 1st, 2nd, 3rd, or 4th liner. The 300+ and 500+ designations are only indicating that for the one graph they were showing all forwards who played 300+ minutes at 5 on 5, while the other shows all forward who played 500+ minutes in all situations. In this particular graph they adjust for zone starts as well (labeled there as "sheltered" vs "buried") to further contextualize the data.

So for example, the graph shows that Luke Glendening gets more minutes than the average fourth liner, but many of those minutes are in the defensive zone. In any case his production (measured by shot generation for Detroit) is pretty bad. Conversely, Tatar gets sheltered third line minutes, but has the best shot generation on the team because of it. Tatar's numbers are impressive because while his icetime indicates that he's roughly in line with a third line player, he's generating more shots than our first and second line guys. His production per TOI is the best on the team. Glendening's is the opposite. He gets the ice time of a 3-4th line tweener, but he struggles.

Share this post


Link to post
Share on other sites

Well you did read it wrong, but you're right that it is badly annotated.

Glendening and Tatar and everybody else's icetime is being compared to NHL line averages to determine whether their TOI is similar to the average 1st, 2nd, 3rd, or 4th liner. The 300+ and 500+ designations are only indicating that for the one graph they were showing all forwards who played 300+ minutes at 5 on 5, while the other shows all forward who played 500+ minutes in all situations. In this particular graph they adjust for zone starts as well (labeled there as "sheltered" vs "buried") to further contextualize the data.

So for example, the graph shows that Luke Glendening gets more minutes than the average fourth liner, but many of those minutes are in the defensive zone. In any case his production (measured by shot generation for Detroit) is pretty bad. Conversely, Tatar gets sheltered third line minutes, but has the best shot generation on the team because of it. Tatar's numbers are impressive because while his icetime indicates that he's roughly in line with a third line player, he's generating more shots than our first and second line guys. His production per TOI is the best on the team. Glendening's is the opposite. He gets the ice time of a 3-4th line tweener, but he struggles.

Your first paragraph is all how I understood it. The part I think is obscure and we're looking at differently is whether they're comparing to a universal average corsi % or if they have an average for 1st line players, a different average for 2nd line, etc. (of course, this would be a scale rather than categories, since players fit between lines on the colour scale).

I think they're doing the later because the of the wording on the bottom axis that says "Corsi for % compared to line average." If so, Glendening's Corsi % (43.9) looks bad on the "all situation" chart because he's getting compared to 2-3 liners since he's purple - who probably don't PK and who's average would be higher than a 4th liner. Maybe 43.9 is good for a player who does a lot of penalty killing and gets all the D zone faceoffs. Even if they're comparing to a league-wide average, it wouldn't tell you that. You would have to compare with other 4th liners - but not based on minutes, based on role or maybe zone starts.

Also, if they're comparing to averages based on line, Tatar's will look inflated on the first chart because he's being compared to 4th liners who probably don't get offensive assignments 5v5. Your way of looking at it shows him ahead of our top line players, but mine says he's being compared to a lower average than those top line players.

Tatar looks good on the "all situations" chart while classified at 3rd line minutes, though, so my reading still ends up with Tatar as very good possession player. Tatar was used in the manner typical to 3rd liners so it would be an apt comparison as opposed to the case with Glen.

Still my biggest takaway from the chart is that Tatar should have gotten much more than 4th line minutes 5v5. I think he's a good 2nd line player.

Edited by PavelValerievichDatsyuk

Share this post


Link to post
Share on other sites

If a 4th line plays as much as a second line, wouldn't that make them a second line?

Not necessarily. It shows that Glendening plays 4th line minutes 5v5. He only has 2-3 line minutes on the all situations graph, but that just says he PKs a lot, since he only got minimal PP time. (He had the 5th most PK time in the league, most of any forward, according to TSN.ca stats).

A regular 2nd line player who probably plays on the PP and has good linemates is not going to have the same corsi as a guy that plays a huge amount on the PK and gets the D zone faceoffs. They both may get 14-15 minutes a game (or whatever it is), but judging shot attempts for the 2 players and being upset that the 2nd player doesn't measure up to the 1st player makes no sense in my view.

Edited by PavelValerievichDatsyuk

Share this post


Link to post
Share on other sites

Well to exclude PK and PP time here are the 5 on 5 stats for the team.

http://stats.hockeyanalysis.com/ratings.php?db=201516&sit=5v5&type=corsi&teamid=11&pos=skaters&minutes=50&disp=1&sort=PCT&sortdir=DESC

CF% = Corsi For / (Corsi For + Corsi Against)

Higher means you shoot more and get shot against less.

Clearly Larkin/Zet/Abdelkader line was struggling. Datsyuk is a possession god.

edit- I shouldn't say they were struggling. 50% is pretty good. Datsyuk/Tatar/Helm just blew it out the water though.

Ironically if you look at last year's the Nyquist/Zet/Abdelkader line was a lot better at around 55% (i got this number from corsica- another source). I think the drop this year is probably because of Z slowing down and not being as dominant on ice.

Edited by kickazz

Share this post


Link to post
Share on other sites

I wanted to go back to a point someone made in another thread about how people who use advanced analytics are acting like they know better than the coaches. This is not true at all. There's things in Blashill's coaching which makes sense and actually works and nobody is refuting the point. The perfect example is Brenden Smith who was sheltered and thrived under Blashill. Another example is demoting Ericsson to third line and promoting Dekeyser to first line. I'd say most of the stuff put out on ice is just fine and obviously Blash and the analytics are lining up in that regard.

The issue comes in where certain MINORITY of the players are being misused and that's where the criticism starts to fall on Blash. Right now there's a few players being misused and those include Glendening, Zetterberg and Drew Miller.

So the next time someone comes about with wacky arguments that "Oh look at the analytics people trying to act like they know better". Stop seeing s*** in black and white, that's not what people are trying to say. They are just pointing out some of the weakest links in the lineup that need to be changed in order to make the team go from being an 8 seed to something better.

"Your team is only as strong as your weakest player"

No the math freaks know everything, that's why they GM there own teams in NHL 16.

Share this post


Link to post
Share on other sites

Soooo...Evander Kane anyone lol?

Sign me up.

No the math freaks know everything, that's why they GM there own teams in NHL 16.

You know that basic stats -- goals, assists, time on ice, +/-, etc. -- are all rooted in mathematics, right? I like that some stats are ok while others HOLY MOTHER OF GOD GET YOUR DEVIL MATHS OUT OF MY HOCKEY YOU SATANIC WIZARD THIS IS A MATH-FREE ZONE!!!

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now