Personally, I don't think most "stats guys" are tracking the game. I think they're using the data presented to them by the few guys that actually do track games live. But I would guess that's a very small minority of people. I mean, why spend all game tracking things that are already being tracked by dedicated websites? This actually supports the claim that a stats guy watches at least as much as anyone else, FWIW.
Player Discussion Neal Pionk: Part II
- Thread starter aufheben
- Start date
- Jul 20, 2005
- 4,890
- 2,253
Not at all. You can be in the correct position doing a good job of keeping your assignment to perimeter play and outside shots and still be in the zone for an entire shift because you cant get a controlled exit. Outside shots are usually less of a problem, but enough of them and you're probably going to see something bad happen.
Think of Girardi at his best. He would go head to head with an opponents top offensive guys and do a great job of limiting what they could actually do in the D-zone. But he spent an eternity defending because the puck never seemed to leave the zone, and when it did it ended up back in the other teams hands.
And that can eventually be a recipe for trouble.
SA16
Sixstring
While it might be a little too simplistic, I agree.
Danny G was a pretty poor defender in transition (esp later on in his career) so not only exiting the zone was a problem but coming in was too easy at times too. I guess it'd be a matter of semantics if gapping up properly and maintaining it in transition would be considered 'positional' play.
But I do agree with the general premise. Girardi 'looked good' in the defensive zone because he was a pretty good risk mitigator outside of those occasional gaffe turnovers like the Ovechkin one and Williams thing in the cup. He limited play to the outside but like you said the volume was crazy high and low danger shots turn dangerous on rebounds and tips etc....so too many wil come back to bite ya.
Pionk probably suffers from the same thing as G did..."looking good" at times by being in position and attacking puck carriers quickly and it is obviously amplified by the offensive flashes like the goal last night.
I think many want to be ‘right’ and hence try to find short cuts and after that you get an endless line of faulty conclusions due to confirmation bias.
There are many that truly believes that Pionk is doing a worthless job out there, due to his rel cf, and it won’t matter if they watch all games for 70 years or not. They are right, CF rel is perfect, and all NHL coaches and assistants and everyone have absolutely no clue what they are doing. This sounds like an exaggeration, but it absolutely is not. If anything it’s an understatement.
If the same crew did not have corsi rel — very very few of them would think that Pionk is as worthless as he is. I know, I was around before corsi existed. Someone like Stralman would be crapped on, not Pionk. So the eye test don’t matter one thing if you don’t know a lot about the game.
For anyone wanting to get an objective opinion of how accurate corsi rel is all you need to do is pick up a site and look at some league numbers — there are so many examples of where it’s all over the place.
It’s to bad because in the end it can be useful. But instead it leads to all the crap where we supposedly would be much better of if we never got NP since it would save management from themselves, because management have found/understood this wonderful tool called corsi rel.
Adam Clendenning sucks so much defensively that he get — extremely —sheltered ice time and hence super star corsi rel. According to the stats he is a super star but he plays 5 games a year. Same with Cody Franson. Cody has been forced to change teams like what 10 times the last handful of years?
will1066
Fonz Drury
- Oct 12, 2008
- 43,951
- 60,159
He blocked shots. A lot of shots. To make up for his lack of speed and positional disadvantages. Lots of shot attempts against that he "suppressed" literally with his body. He and his career started aging at a presidential rate in more recent years. I think Pionk's story is that he stick checks a lot, lots of chipping the puck ahead to his defensive partner or a forward who is supporting, and using his escape speed to clear the puck out of his own end.
Larrybiv
We're CLEAN, we PROMISE!
I would like to thank nyk2k as he was my 500th like.....and to all of those who have "liked" some of my comments over the years. Thanks.
eco's bones
Registered User
FWIW that is what John Tortorella wanted his defenseman to do and Girardi became a shot blocking machine and that beat him up and hurt the rest of his game. AV then comes along with his man on man defensive system which played to Girardi's weaknesses which had been exacerbated because of Tortorella's love for shot blocking. So then Girardi goes to Tampa--a bunch of posters from here go to warn posters from there how terrible Girardi is and then Girardi has a pretty good year there because Jon Cooper doesn't put him into situations that play to his weaknesses which makes the Tampa posters wonder what the posters here were talking about. They like him just fine. Girardi's not that far away from 1000 NHL regular season games--he's got well over 100 playoff games. I'm glad he's in Tampa instead of here but he's a good guy and a decent player if you use him correctly.
eco's bones
Registered User
But anyway on Pionk I like him. The Rangers have been overusing him and playing him a lot against other team's better players. The way DeAngelo is playing the Rangers could see if he has the engine to play as many minutes but not a lot of defenseman have the energy to play 25 minutes a night for very long without burning out. Pionk seems to be one who can. I mean--here he is last night making an end to end rush practically at the last minute.
Not to get me wrong DeAngelo has kind of been a revelation this year in his own right. His defense looks a lot better this year--his overall games looks a lot better.
Not to get me wrong DeAngelo has kind of been a revelation this year in his own right. His defense looks a lot better this year--his overall games looks a lot better.
Goddam what a goal.
Doesn’t make up for him getting absolutely dominated 90% of the time he’s on the ice, but her certainly does some impressive stuff the other 10% of the time. He plays exactly like the protoypical offense only defenseman everyone complains about.
Doesn’t make up for him getting absolutely dominated 90% of the time he’s on the ice, but her certainly does some impressive stuff the other 10% of the time. He plays exactly like the protoypical offense only defenseman everyone complains about.
darko
Registered User
- Feb 16, 2009
- 70,268
- 7,796
I think it matters a lot more than everyone thinks. And the counterary have definitely NOT been proven.
Look this is the problem, if I can call it ‘you guys’, never critically look at these claims.
You want to criticize the use of +\~. Due to that so many can point at some very solid factors for why its pretty meaningless to compare two players rating in that column. But name one point of critique against +\- that doesn’t apply to corsi, besides the fact that plus minus includes shorties. One.
It can’t be done, the only difference between +\~ and corsi is that corsi takes away quality of goaltending and shot blocking, but you can also argue that +\~ has an advantage over corsi since it’s defacto counts goals and hence to a large extent quality of shot.
And, the icing on the cake, goal difference has the last years been a better teller of success than corsi. And surprise surprise, Pionk is at the bottom of our team in +\~ too.
My point is just, can we just admit that corsi more or less the same thing as +/-? Marek Malik can lead the league in +\- and Patrick Weinroch can lead the league in corsi. Both have a margin of error the size of the Pacific if you believe that they stand alone can be looked at.
Corsi beats out +/- since you need a smaller sample size that is for sure. But still — the exact same point of criticism against +\- applies on corsi besides that shorties count as a minus.
How many knows that you can produce a player’s exact corsi by just taking his +\- and counting backwards if you have shot attempts going to net and sv% for both teams when that player is on the ice (and of course number of shorties scored that must be removed)?
And those two factors, shot attempts going to the net and sv%, basically only differ a few percents between the top and the bottom team in the league? It’s not like one team has a goalie that saves 80% of the shots and another 95%. The difference is maybe handful of percent. And it’s the same with shot attempts finding the net.
And those two factors, shot attempts going to the net and sv%, basically only differ a few percents between the top and the bottom team in the league? It’s not like one team has a goalie that saves 80% of the shots and another 95%. The difference is maybe handful of percent. And it’s the same with shot attempts finding the net.
Irishguy42
Mr. Preachy
It's not that QoC doesn't matter. It's just that QoT matters wayyyyyyyyy more, and QoC does end up becoming a wash over the case of a full season.
Distribution of Quality of Competition and Teammates Metrics
Why Quality of Competition doesn’t matter to analytics experts anymore
This is proven.
Distribution of Quality of Competition and Teammates Metrics
Why Quality of Competition doesn’t matter to analytics experts anymore
This is proven.
@ManUtdTobbe I am rounding this off at three posts, don’t worry, but one last thing.
A statement that QoO don’t matter is in itself 100% illustrative of the level of this debate. 6 players on one side plays 6 players on the other side. Both sides tries to score the most goals. At the end you count the result.
Don’t you find it odd that QoO wouldn’t matter in that perspective? It’s in fact all that matters. If you can’t see it in the data, it’s because it’s clouded.
A statement that QoO don’t matter is in itself 100% illustrative of the level of this debate. 6 players on one side plays 6 players on the other side. Both sides tries to score the most goals. At the end you count the result.
Don’t you find it odd that QoO wouldn’t matter in that perspective? It’s in fact all that matters. If you can’t see it in the data, it’s because it’s clouded.
Irishguy42
Mr. Preachy
Larrybiv
We're CLEAN, we PROMISE!
Irishguy42
Mr. Preachy
Ouch, you are making such a big mistake if you take the conclusions in that article and believes that QoC don't matter that much for all individual players.
Stay with me a bit on this, please. Theoretically, QoC matters as much as quality of linemates. Right? Two cars are racing, the speed of both cars matters the same. So why does the numbers in that article show that in the NHL when you look at the data from all players during an entire season -- quality of linamates matter more than quality of competition? Its of course pretty simple, you will of course have much more effective ice time with your linemates than the effectiveness of the ice time against your opponents. Every game all players log ice time against all opponents. Players are left on the ice after shifts, hops on the ice first, gets on the ice and are changed ASAP to get the right match-up and so forth. One team dumps the puck in and a defender retrieves it while 8-9 other skaters are changing.
Of a 40 sec shift, I would bet that on average you often got 10-15 sec that are completely meaningless time as regards a players quality. In other words, if you really scrutinized a game by Sidney Crosby in which he plays 14 minutes of 5 on 5 ice time, how big portion of those 14 minutes could you realistically say that Crosby actively impacting the play on the ice? You can remove all the time he is skating onto the ice from the bench and off the ice after a shift. That is what like 8 seconds per shift that doesn't end or start with an whistle. In addition you have all the in game situations during which he would have zero impact, all the dead time after a dump in. The last seconds of a period. You have all those shifts were a top unit plays 30-40 seconds against its match-up unit, then is the opponents changed and the top unit is pinned in their own end by the fresh line they face (i.e. the non-matchup unit face high quality opponents at the time the high quality opponents are more vunerable more often than match-up units). And so forth and so forth.
So if Crosby plays 14 minutes of 5 on 5 ice time during a game, a big portion of that ice time will not be effective. And of the ineffective ice time, the biggest portion of it will of course come against players that he is -- NOT matched against.
So what that article shows is de facto just how ineffective a lot of the ice time is, i.e. how unreliable the raw data is.
Then you might ask, why haven't I seen 25 posts on Twitter point the above out if its this easy? Because nobody posting on this topic on twitter are looking to critically question metrics in areas where you have no obvious solution.
IMO the only sane conclusion you can make is that QoC matters more for match-up units whilst of course, to an equal extent, less for everyone else.
Last edited:
I'm going through your history and unliking all your comments that I previously liked
