Statistics can be interesting

[RWMc1]

21 minutes ago, oldnews said:

 

What is the correlation between "soft" and giveaways?

Isn't it obvious. If an opposing player pokes your stomach and you give the puck away while you do the doughboy laugh, it means you're soft. Have you never played hockey?

[oldnews]

23 minutes ago, RWMc1 said:

Isn't it obvious. If an opposing player pokes your stomach and you give the puck away while you do the doughboy laugh, it means you're soft. Have you never played hockey?

Duh - some excessive derp level there considering you've come up with such a patronizing and oversimplified response lulz.

 

Giveaways do not categorically equal 'softness'.  That much should be obvious.

 

Burns, Seabrook, Chara....must be 'soft', right.

[Googlie]

OK - it's been postulated that heavy TOI correlates to giveaways.

That explains Edler - most TOI (24.37 per game), but not Henrik (7th, excluding Fedun, at 18.35). But shouldn't Daniel be up there too  ....8th at TOI, 18.21, but 9th on giveaways, at 18 total.  Is this accounted for by Henrik being a center, passing more, with more stray passes?

And if this correlation indeed exists, then Hutton is a lost cause. 10th on TOI (at 18.13 per game) but 5th on giveaways (24 total)

*****************************************************************************************************************************************************************

Maybe the ratio of takeaways to giveaways is a better measure of a player's performance in making the right decisions re puck movement.  Here (again, only those with 10+ games) our perceptions are not supported by the statistics.  The worst (at 3.3 times more giveaways that takeaways), are Tanev and Prust (and not Edler!), followed by, at 2 1/2 more giveaways than takeaways, by Sbisa, Hamhuis, Edler and Bartowski.

The best ratios are held by Burrows (3 x more takeaways than giveaways), Virtanen and Cracknell (1.8 times), Horvat at 1.3, then Hansen and McCann at 1.2 times.

So: Prust is a liability, while Alex "trade him he's over the hill" Burrows is an asset. Virtanen doesn't hurt the team. How the hell did Horvat manage to get to -23?

 

 

[RWMc1]

15 minutes ago, oldnews said:

Duh - some excessive derp level there considering you've come up with such a patronizing and oversimplified response lulz.

 

Giveaways do not categorically equal 'softness'.  That much should be obvious.

 

Burns, Seabrook, Chara....must be 'soft', right.

That says it all. It should have been plainly obvious that I was being facetious. Sorry, that means I was joking.

[oldnews]

10 minutes ago, Googlie said:

OK - it's been postulated that heavy TOI correlates to giveaways.

That explains Edler - most TOI (24.37 per game), but not Henrik (7th, excluding Fedun, at 18.35). But shouldn't Daniel be up there too  ....8th at TOI, 18.21, but 9th on giveaways, at 18 total.  Is this accounted for by Henrik being a center, passing more, with more stray passes?

And if this correlation indeed exists, then Hutton is a lost cause. 10th on TOI (at 18.13 per game) but 5th on giveaways (24 total)

*****************************************************************************************************************************************************************

Maybe the ratio of takeaways to giveaways is a better measure of a player's performance in making the right decisions re puck movement.  Here (again, only those with 10+ games) our perceptions are not supported by the statistics.  The worst (at 3.3 times more giveaways that takeaways), are Tanev and Prust (and not Edler!), followed by, at 2 1/2 more giveaways than takeaways, by Sbisa, Hamhuis, Edler and Bartowski.

The best ratios are held by Burrows (3 x more takeaways than giveaways), Virtanen and Cracknell (1.8 times), Horvat at 1.3, then Hansen and McCann at 1.2 times.

So: Prust is a liability, while Alex "trade him he's over the hill" Burrows is an asset. Virtanen doesn't hurt the team. How the hell did Horvat manage to get to -23?

 

 

Edler - and any other big minute blueliner who plays hard minutes - has the task of moving the puck in the most challenging of circumstances - ie with forecheckers in his face - attempting to initiate breakouts or simply avoid being separated from the puck by pending hits, etc.  He also winds up possessing and moving the puck a great deal - the sheer volume alone leads to a fair amount of inevitable turnovers.

The Sedins - like most forwards - simply don't handle the puck in as precarious situations - and are elite - as good as anyone ever has been - at completing their passes.

Ratio of turnovers doesn't complete the picture either.  Horvat for example is playing the hardest minutes of any forwards - lots of heavy lifting down low in his own zone attempting to support his (depleted at times) blueline - more integral to the difficult breakouts with heavy forechecking bearing down on him - and top line forwards looking to maintain ozone possession.  Playing those shutdown minutes also puts him in a lot of high risk (of turnovers) situations. 

I think you may be attempting to read too much into isolated turnover and turnover differential alone.

[Dekey Pete]

One or two stats (or simply examining a ratio in your case) often don't paint a complete picture of the situation.  Stats are useful to tell a story, but to compare someone's stat X against another person's stat X without the analysis to support it you can't come to any meaningful conclusion.

[Quoted]

54 minutes ago, Googlie said:

OK - it's been postulated that heavy TOI correlates to giveaways.

That explains Edler - most TOI (24.37 per game), but not Henrik (7th, excluding Fedun, at 18.35). But shouldn't Daniel be up there too  ....8th at TOI, 18.21, but 9th on giveaways, at 18 total.  Is this accounted for by Henrik being a center, passing more, with more stray passes?

And if this correlation indeed exists, then Hutton is a lost cause. 10th on TOI (at 18.13 per game) but 5th on giveaways (24 total)

*****************************************************************************************************************************************************************

Maybe the ratio of takeaways to giveaways is a better measure of a player's performance in making the right decisions re puck movement.  Here (again, only those with 10+ games) our perceptions are not supported by the statistics.  The worst (at 3.3 times more giveaways that takeaways), are Tanev and Prust (and not Edler!), followed by, at 2 1/2 more giveaways than takeaways, by Sbisa, Hamhuis, Edler and Bartowski.

The best ratios are held by Burrows (3 x more takeaways than giveaways), Virtanen and Cracknell (1.8 times), Horvat at 1.3, then Hansen and McCann at 1.2 times.

So: Prust is a liability, while Alex "trade him he's over the hill" Burrows is an asset. Virtanen doesn't hurt the team. How the hell did Horvat manage to get to -23?

 

 

Those stats can't really be compared between forwards and defense though.

 

[Kragar]

58 minutes ago, Googlie said:

OK - it's been postulated that heavy TOI correlates to giveaways.

That explains Edler - most TOI (24.37 per game), but not Henrik (7th, excluding Fedun, at 18.35). But shouldn't Daniel be up there too  ....8th at TOI, 18.21, but 9th on giveaways, at 18 total.  Is this accounted for by Henrik being a center, passing more, with more stray passes?

And if this correlation indeed exists, then Hutton is a lost cause. 10th on TOI (at 18.13 per game) but 5th on giveaways (24 total)

*****************************************************************************************************************************************************************

Maybe the ratio of takeaways to giveaways is a better measure of a player's performance in making the right decisions re puck movement.  Here (again, only those with 10+ games) our perceptions are not supported by the statistics.  The worst (at 3.3 times more giveaways that takeaways), are Tanev and Prust (and not Edler!), followed by, at 2 1/2 more giveaways than takeaways, by Sbisa, Hamhuis, Edler and Bartowski.

The best ratios are held by Burrows (3 x more takeaways than giveaways), Virtanen and Cracknell (1.8 times), Horvat at 1.3, then Hansen and McCann at 1.2 times.

So: Prust is a liability, while Alex "trade him he's over the hill" Burrows is an asset. Virtanen doesn't hurt the team. How the hell did Horvat manage to get to -23?

 

 

Also, total TOI, not TOI per game, should be used.  If you are looking at total giveaways/takeaways, they need to be compared against total TOI.

If you have someone getting 5 shifts a game give the puck away once, and someone else getting 20 shifts a game giving it away twice, who is the more reliable player from a giveaway standpoint?

[JamesB]

2 hours ago, SID.IS.SID.ME.IS.ME said:

Have a look two posts above your own.

James B (+1) breaks things down pretty much the same way I would, regarding high raw giveaway totals and what they actually say about performance.

EDIT: And as far as management goes, I'm sure they are well aware of all this.

I highly doubt that Benning really believes that low raw giveaway totals are indicators of good performance. But he likely tasked Jonathan Wall (Canucks analytics guy) for some positive sounding stats on Sbisa that he could trot out during that presser.

And for a while at least, it worked as he intended and we saw lots of people quoting JB and suggesting that being 5th on the D in giveaways made Sbisa a "real good" player last year.

By the time we saw the backlash from the more stats-savvy community, the fair weather fans had already moved on to the next story. All in all, a nice piece of propaganda/PR work by management.

Nice analysis of the PR aspects of Benning's comments. I had not thought of that but it makes a lot of sense. On reflection, I remember Benning talking about Sbisa's giveaway numbers and it did kind of look like he was repeating something rehearsed that he did not actually care about.

And I am sure you are right that Benning understands the implications of giveaway numbers.

But I would still feel better if Benning paid more attention to using analytics in salary negotiations. I admit that Benning seems to be very good as assessing young talent based on the "eye test". For guys in Junior, the NCAA, Europe, and even the AHL, there is not nearly as much reliable performance data as for the NHL, so subjective evaluations are crucial, and Benning seems to be one of the best in the business in this area. At the NHL level, however, analytics can give some guidance and I think it does seem like Benning has over-estimated the value of several NHL players, including Dorsett, Prust, Sbisa, Miller, Vrbata, and Sutter. I am not saying they are bad players. They are just getting paid a lot given their underlying analytics.  And it is not clear that their "intangibles" make up for it.

[Googlie]

2 hours ago, Kragar said:

Also, total TOI, not TOI per game, should be used.  If you are looking at total giveaways/takeaways, they need to be compared against total TOI.

If you have someone getting 5 shifts a game give the puck away once, and someone else getting 20 shifts a game giving it away twice, who is the more reliable player from a giveaway standpoint?

I did look at totals, but then discrepencies enter into it for someone playing 15 games vs 25 vs 45, etc.

Maybe a better determinant would be giveaways per game modified by avg TOI (ie giveaways per 10 mins, or something like that).  Could also see if correlation between giveaways per shift, but then that doesn';t account for length of shift (eg, do most of Henrick's giveaways occur at the end of long shifts?

And I wonder if NHL keeps track of icings per player?

[Googlie]

5 hours ago, Dral said:

 

What do giveaways have to do with being soft?

In one of the other threads, someone said that Vey played soft, gave away the puck as soon as he was nudged. That led me to look at the giveaway stats.

[TOMapleLaughs]

Here's an interesting stat:  

(Well, standings more than stats.)

The Canucks currently have just two more wins than the dead-last Columbus Blue Jackets.

[Kragar]

1 minute ago, Googlie said:

I did look at totals, but then discrepencies enter into it for someone playing 15 games vs 25 vs 45, etc.

Maybe a better determinant would be giveaways per game modified by avg TOI (ie giveaways per 10 mins, or something like that).  Could also see if correlation between giveaways per shift, but then that doesn';t account for length of shift (eg, do most of Henrick's giveaways occur at the end of long shifts?

And I wonder if NHL keeps track of icings per player?

Right, that's what I was getting at.  Vey only playing 15 games so far should mean his giveaway totals are lower.  Virtanen has played more games, but with less TOI, and likely less opportunity to handle the puck (center vs winger), so his slightly better total than Vey doesn't make him a better puckhandler.  Your comment about shift length adds is a good point, btw.

Here's another: how often they handle the puck.  As some have mentioned, Edler is in a PMD role, and Henrik is a center known for his passing skills.  Top players are going to get the puck more, and therefore have more opportunities to lose the puck along with having more opportunity to score.  Virtanen and Vey do not have the same opportunities that Edler and Henrik do.

That's part of the reason why stats only get you so far.  They often don't tell the whole picture. 

No idea on tracking icings.  You might see similar numbers there though... the more reliable defensive-type players are put into tougher d-zone situations, and may be more likely to ice it than someone who is not called on to defend against the opposition's top lines, but that doesn't mean they are necessarily worse defenders.

[Davathor]

9 hours ago, Googlie said:

grrrr

Content didn't post

Intended to say ......

Perceptions often differ from reality

for example, posters have ragged on Vey for being soft. yet among players with 10 or more games, he is second best on giveaways, with 5 (Virtanen leads in fewest, at 4)

Worst is Edler (at 40) bur a surpising 2nd worst is Henrik, at 31

Similarly with takeaways.

Surprising first is McCann, at 21, followed by Burr and Cracknell at 18

Who'd a thought it?

 

Vey has played less games than almost everyone, of course he's going to have the least giveaways

 

Books can be interesting, but not if you can't understand them 

[Googlie]

Biggest surprise for me was McCann - among regular forwards, 3rd lowest time on ice per game (only Baertschi and Jake with fewer minutes), but leads the team in takeaways.  He is also middle of the pack (7 out of 14 with 10+ games) among forwards in blocking shots.  He just needs to work on his face-offs

If he doesn't pan out as a #1 Center, there might be a future as a 4th line defensive center

[Baggins]

5 hours ago, Googlie said:

Biggest surprise for me was McCann - among regular forwards, 3rd lowest time on ice per game (only Baertschi and Jake with fewer minutes), but leads the team in takeaways.  He is also middle of the pack (7 out of 14 with 10+ games) among forwards in blocking shots.  He just needs to work on his face-offs

If he doesn't pan out as a #1 Center, there might be a future as a 4th line defensive center

That's because he can't win a faceoff. He's always chasing the puck.

[ice orca]

9 hours ago, Googlie said:

Biggest surprise for me was McCann - among regular forwards, 3rd lowest time on ice per game (only Baertschi and Jake with fewer minutes), but leads the team in takeaways.  He is also middle of the pack (7 out of 14 with 10+ games) among forwards in blocking shots.  He just needs to work on his face-offs

If he doesn't pan out as a #1 Center, there might be a future as a 4th line defensive center

Another thing about McCann that Botchford pointed out, the Canucks have more shots on goal when McCann plays vs when he doesn't play. The kid does have an active stick and good positioning and thats probably why he is good at turning the puck over.

[ice orca]

3 hours ago, Baggins said:

That's because he can't win a faceoff. He's always chasing the puck.

Yeah that's his weak spot but he will get better, you can see that in his personality and compete level even at 19.

[Hortankin]

On 26/01/2016 at 1:38 PM, Googlie said:

grrrr

Content didn't post

Intended to say ......

Perceptions often differ from reality

for example, posters have ragged on Vey for being soft. yet among players with 10 or more games, he is second best on giveaways, with 5 (Virtanen leads in fewest, at 4)

Worst is Edler (at 40) bur a surpising 2nd worst is Henrik, at 31

Similarly with takeaways.

Surprising first is McCann, at 21, followed by Burr and Cracknell at 18

Who'd a thought it?

More minutes = more giveaways/takeaways

 

Really not that interesting lol

[Googlie]

40 minutes ago, McHortanen said:

More minutes = more giveaways/takeaways

 

Really not that interesting lol

so Edler, Tanev, Daniel, Bart and Bo should be our 5 leaders in giveaways/takeaways.

But here's the thing ..... they're not.  On giveaways, the 3 defensemen are, but they are joined by Henrik and Vrby, not the 2 forwards with the most ice time. On takeaways, only Bo is in the top 5.

Which contributes to the head-scratcher .... why is Bo -24?  he has 5th most TOI,  he is in the top 3rd in fewest giveaways among regular forwards, is among top 5 in takeaways, 2nd best forward in blocking shots, top 10 in hits. He's top 5 in goals and top 5 in assists.

He's taken most face-offs (900) winning 50.6 of them (Only Sutter and Prust !! are better)  He's 6th in shots on goal, and seventh in conversion %

This all adds up to a pretty formidable competitor, which of course he is,  But he has the worst +/- on the team, But not only that, he has the worst +/- in the league!!

What gives?