News Article: The Pens' Improved Possession Numbers: A Product of Health or a Wretched Opponent?

[Crafton]

Pittsburgh is a team that was seen as a bit of a paper tiger by a lot of numbers guys heading into the playoffs because they were into their second year of abysmal possession numbers when Sidney Crosby or Evgeni Malkin weren’t on the ice.

Why has this happened? There are two explanations that seem most likely. First, the Penguins are finally healthy after a tough season and have gelled at the right time. I’m skeptical that this is true. Their problems with depth extend back to last season. What seems more likely to me is that the Rangers are just in a tough spot in the schedule in terms of games played. Pittsburgh wrapped up their first round series April 28, the Rangers didn’t finish their series until April 30.

http://www.sportsnet.ca/hockey/nhl/pens-depth-enough-to-beat-tired-rangers/

the second year in a row bit is interesting. what of Letang and Martin's effect on the numbers? their health could be the key to the penguins ability to possess the puck.

Adam Gretz ‏@AGretz Apr 9
The Penguins have had Kris Letang and Paul Martin in the lineup just 96 times the past 3 years. They're 64-25-7 and a 53 percent FenClose.

https://twitter.com/AGretz/status/453907884330741760

lord knows that skepticism of this team is warranted, but does this article give the team too little credit? too much credit?

Click to read more...

 

[WVP]

Both seems to be the obvious answer.

 

[Crafton]

Both seems to be the obvious answer.

indubitably

 

[sf expat71]

Their return and Orpik's injury also reduced Scud's minutes as well. I truly think his abilities as a possession drain cannot be overstated.

 

[WVP]

Also, the Malkin-Crosby combo is working very well for the moment. Neal was a ghost with Malkin but is competing more with Sutter. Likewise both Crosby and Malkin are getting high quality scoring chances together.

It's obviously not optimal for the long term but is looking like our best configuration this spring.

 

[JQR]

"Getting to our game."

 

[AjaxTelamon]

Their return and Orpik's injury also reduced Scud's minutes as well. I truly think his abilities as a possession drain cannot be overstated.

Completely agree, you could not find a larger swing in possession than swapping in Letang for Orpik on our top minute-eating D pair. And as others have mentioned, having your top 2 D pairings loaded with PMD's cannot be understated either.

 

[HandshakeLine]

And why exactly does the author discount the "gel'd at the right time" theory? And, furthermore, why doesn't said author consider other potential causes for post-season success, such as heathy/scratched players, match-ups, etc?

Sports journalism is literally the worst thing, written by the worst people this side of a maximum security prison and/or politics.

edit: After reading an excerpt posted on my brothers blog, what I've read smacks of writing the conclusion before the article. And of course, it's on Sportsnet.

 

[cygnus47]

Because gelling as a team is something that is always predicted by smoother upswings in stats. We've basically gone from being rubbish for four months straight, to suddenly possessing the puck at like 70% outside of game 3. That is a huge outlier in terms of statistical trends which suggests that it's more than just the team improving.

 

[wgknestrick]

It is 80% personnel. Remaining 20% is player effort (mostly Sutter) and different line combos.


FF%

Guys out:

Orpik 47%
Glass 41%
Pyatt 41%
Engo 45%
Sutter (regular season) 44%
Adams (regular season) 43%

Guys in (playoff FF%s so far):

Letang 55%
Gibbons 56%
BB 51%
Goc 56%
Sutter (playoffs) 48%
Adams (playoffs) 50%

Are we still wondering what changed? A hockey lineup is like a poker hand. It is wise to throw away your worst cards for better ones (what this board has been asking for a while now). I may have some of the replacements off, but I am darn close to showing where all the improved possession came from.

 

[KIRK]

http://www.sportsnet.ca/hockey/nhl/pens-depth-enough-to-beat-tired-rangers/

the second year in a row bit is interesting. what of Letang and Martin's effect on the numbers? their health could be the key to the penguins ability to possess the puck.



https://twitter.com/AGretz/status/453907884330741760

lord knows that skepticism of this team is warranted, but does this article give the team too little credit? too much credit?

Sid and Geno together. The cohesion of the other lines, with guys playing to their relative strengths. Having two puck possession defensive pairings getting the lions share of minutes and limited ES minutes for Scuds. Better commitment to defense leading to more takeaways. Better breakouts. And, yes, the Rangers are an ideal matchup.

I'll put it this way: It wouldn't look THIS good if we weren't playing the Rangers, but it will look better than it has in the past against whoever we face next round if we change nothing in terms of who plays and who plays with whom.

 

[JQR]

Sid and Geno together. The cohesion of the other lines, with guys playing to their relative strengths. Having two puck possession defensive pairings getting the lions share of minutes and limited ES minutes for Scuds. Better commitment to defense leading to more takeaways. Better breakouts. And, yes, the Rangers are an ideal matchup.

I'll put it this way: It wouldn't look THIS good if we weren't playing the Rangers, but it will look better than it has in the past against whoever we face next round if we change nothing in terms of who plays and who plays with whom.

I'm actually considering not watching the first game of the next round. For obvious reasons.

 

[KIRK]

I'm actually considering not watching the first game of the next round. For obvious reasons.

One question: Expecting the meltdown before or during?

 

[JQR]

One question: Expecting the meltdown before or during?

I'm expecting Bylsma lines out of the gate, and I really really really want to avoid seeing that. Even when he inevitably switches back to what is working, I will still be pissed.

 

[KIRK]

I'm expecting Bylsma lines out of the gate, and I really really really want to avoid seeing that. Even when he inevitably switches back to what is working, I will still be pissed.

 

[JQR]

 

[Freeptop]

http://www.sportsnet.ca/hockey/nhl/pens-depth-enough-to-beat-tired-rangers/

the second year in a row bit is interesting. what of Letang and Martin's effect on the numbers? their health could be the key to the penguins ability to possess the puck.



https://twitter.com/AGretz/status/453907884330741760

lord knows that skepticism of this team is warranted, but does this article give the team too little credit? too much credit?

The second year in a row bit neglects the fact that the Pens dealt with some significant injuries last season, too.
Crosby, Malkin, Martin and Letang all missed significant portions of last season. Guess which four players have the biggest impact on puck possession on the team?

It's also worth noting that having those guys in the game cause a trickle-down effect that aids the possession numbers of other players, since players like Sutter get to face lesser competition.

We'll see how things go from here, but I think it's giving too little credit to the Pens to say their improvements have come solely due to the Rangers' fatigue. There has been a marked difference in the way the Pens have played the last few games, and the way they've played ties directly to their improvements in puck possession statistics. Improved health, and improved buy-in for Jacques Martin's defensive schemes should get more credit than Tyler Dellow was willing to give them. Of course, there's also the fact that the Rangers' own possession stats were boosted by playing the Flyers in the first round (the Flyers were 23rd in FenwickClose% during the regular season).

Speaking of opponents' possession stats, the Blue Jackets actually had a higher FenwickClose% during the regular season than the Pens did (albeit, not by much), but the Pens outpossessed the Blue Jackets in round 1 (Blue Jackets' FenwickClose% was 42.3% in Round 1). Considering that fatigue wasn't an issue for the Blue Jackets, that would seem to indicate that there were improvements from the Pens' end.

Don't get me wrong - I do think the Rangers' fatigue does play into the Pens' numbers being as ridiculously good as they have been. But Dellow seems to want to discount anything other than the Rangers' fatigue as being a factor.

I do find it amusing that one of the big proponents of "advanced stats" is now trying to come up with reasons as to why we shouldn't pay attention to the Penguins' advanced stats right now, though. After all, the Pens' FenwickClose% and PDO make them look really good right now.

 

[Shady Machine]

Pretty poor logic in the conclusion. Basically saying the schedule is the sole reason for the turn around. That says nothing for the line up changes, obvious change in approach with minor systematic changes, Pen's historical success against the Rangers, etc.

The schedule and injuries to key Ranger players (assuming Pierre isn't just exaggerating) is probably an issue, but not enough to ***** about it after they lose.

 

[wgknestrick]

The second year in a row bit neglects the fact that the Pens dealt with some significant injuries last season, too.
Crosby, Malkin, Martin and Letang all missed significant portions of last season. Guess which four players have the biggest impact on puck possession on the team?

It's also worth noting that having those guys in the game cause a trickle-down effect that aids the possession numbers of other players, since players like Sutter get to face lesser competition.

We'll see how things go from here, but I think it's giving too little credit to the Pens to say their improvements have come solely due to the Rangers' fatigue. There has been a marked difference in the way the Pens have played the last few games, and the way they've played ties directly to their improvements in puck possession statistics. Improved health, and improved buy-in for Jacques Martin's defensive schemes should get more credit than Tyler Dellow was willing to give them. Of course, there's also the fact that the Rangers' own possession stats were boosted by playing the Flyers in the first round (the Flyers were 23rd in FenwickClose% during the regular season).

Speaking of opponents' possession stats, the Blue Jackets actually had a higher FenwickClose% during the regular season than the Pens did (albeit, not by much), but the Pens outpossessed the Blue Jackets in round 1 (Blue Jackets' FenwickClose% was 42.3% in Round 1). Considering that fatigue wasn't an issue for the Blue Jackets, that would seem to indicate that there were improvements from the Pens' end.

Don't get me wrong - I do think the Rangers' fatigue does play into the Pens' numbers being as ridiculously good as they have been. But Dellow seems to want to discount anything other than the Rangers' fatigue as being a factor.

I do find it amusing that one of the big proponents of "advanced stats" is now trying to come up with reasons as to why we shouldn't pay attention to the Penguins' advanced stats right now, though. After all, the Pens' FenwickClose% and PDO make them look really good right now.

You just have to realize that their advanced stats "now" have been playing against only 2 teams.....not the entire NHL......and the "now" stats are only over 10 games. You don't want to get too confident about them because they haven't been playing top competition (like CHI/BOS/LA) and still getting those numbers.

I still attribute Sutter's play as the biggest factor at improving other players stats. Sutter has been given just about every defensive zone start that is available AND he has been able to move the puck into the Ozone....all while outshooting the other team.

This cascades to other players because they don't have to start in the D zone (think how GlAdams shifts used to end), and they can focus on getting more shots for. Sutter has been setting up the entire team for success because he has been eating up the Dzone starts (and is still productive himself), while winning the territorial war.

 

[wgknestrick]

That study does not paint a pretty picture for Glass. His benching seems to have had a significant impact on Goc.

 

[SaintLouHaintBlue]

This may be a bit off topic, but -
Is it just me, or are corsi & fenwick numbers not really useful for anything except to show that a particular individual player maybe...probably sucks?

Corsi & Fenwick are basically fancy +/-, except with shots, not just GF/GA.

I really couldn't fathom using any advanced stats in hockey to assemble a team, the way you might be able to in baseball (sabermetrics, Moneyball).

 

[Malkin4Top6Wingerz]

This may be a bit off topic, but -
Is it just me, or are corsi & fenwick numbers not really useful for anything except to show that a particular individual player maybe...probably sucks?

Corsi & Fenwick are basically fancy +/-, except with shots, not just GF/GA.

I really couldn't fathom using any advanced stats in hockey to assemble a team, the way you might be able to in baseball (Sabremetrics, Moneyball).

You can't use them in the same way as baseball because there are many, many more variables at play. Fortunately many of those have methods of being quantified as well (quality of linemates / opposition, zone starts, etc.), so we're getting closer to being able to break down a player's worth in a similar kind of way. It will never be as clean as baseball, though, and it's likely more useful for judging teams than individual players. Still, it's a valuable tool when viewed in the right context and should be viewed as such.

 

[wgknestrick]

This may be a bit off topic, but -
Is it just me, or are corsi & fenwick numbers not really useful for anything except to show that a particular individual player maybe...probably sucks?

Corsi & Fenwick are basically fancy +/-, except with shots, not just GF/GA.

I really couldn't fathom using any advanced stats in hockey to assemble a team, the way you might be able to in baseball (Sabremetrics, Moneyball).

They are very, very useful.

#1 They show where the player is spending most of his time when on the ice. + corsi = in the Ozone, - corsi = in his own D zone. Obviously you want players that can move the play to the other team's zone. Ie anti Tanner Glass. You can't outscore the other team consistently if you are always losing the shot battle.

#2 A player's zone starts are an important aspect of context with the stats. If a coach usually sends a player out for D zone faceoffs, he will tend to have a lower corsi because he is starting territorially where the other team can shoot immediately. This is the largest "context" factor to remember. This is also why Sutter's play is so great. He is doing very well corsi wise and goal wise (in light of him starting a very high # of shifts in his D zone). Another smaller factor to consider is what kind of players does the player usually play against, positive corsi players/or negative? This is Qual Comp. Once again, a factor, but the smallest on Corsi. Generally you will see players with low shooting %s, start shifts in their own zone. Players with high SH% will most likely start in the other team's zone.....assuming a coach is smart. Since a high SH% player is more likely to get a shot-for in the O zone, you start them there to maximize your goals for.

#3 Corsi is a better indicator than "goals" (ie +-) because they aren't as rare of an event. You get about 10x more shot events than goal events. The more data you base your decision on, the better. While corsi ignores "shot quality", most players are under the meat/mean of the bell curve in "shot quality". It is a factor that you can usually ignore for most players (outside of TK or Crosby who are at the extremes of shot quality-for populations).

 

[Freeptop]

You just have to realize that their advanced stats "now" have been playing against only 2 teams.....not the entire NHL......and the "now" stats are only over 10 games. You don't want to get too confident about them because they haven't been playing top competition (like CHI/BOS/LA) and still getting those numbers.

I still attribute Sutter's play as the biggest factor at improving other players stats. Sutter has been given just about every defensive zone start that is available AND he has been able to move the puck into the Ozone....all while outshooting the other team.

This cascades to other players because they don't have to start in the D zone (think how GlAdams shifts used to end), and they can focus on getting more shots for. Sutter has been setting up the entire team for success because he has been eating up the Dzone starts (and is still productive himself), while winning the territorial war.

Oh, I agree that the sample size is too small from which to draw conclusions about the Pens in potential upcoming rounds. In fact, I think the basic premise of the article, which is looking at the numbers and asking why are they so different from the regular season is reasonable. I'm just pointing out that singling out one factor and proclaiming it to be the sole one that matters seems more like ignoring the greater context, rather than being a useful dive in to the analytics.

The irony is, I went to Tyler Dellow's site and found this article:
http://www.mc79hockey.com/?p=6949
In this article, he writes about how changes in how the Blue Jackets used their personnel made a huge difference in their possession metrics from the start of the season compared to the end, so I found it really odd that he made no attempt to look at how the Penguins made different use of their personnel. He just jumped straight to pointing at the Rangers' fatigue.

Compare these two conclusions about similar circumstances:

In any event, if you’re looking for what changed when Columbus erupted following December 1, there’s your answer. A large part of their roster took off. It would be interesting to hear from some Columbus watchers as to whether there were big tactical changes or anything with the team – I’m inclined to suspect that there were.

Why has this happened? There are two explanations that seem most likely. First, the Penguins are finally healthy after a tough season and have gelled at the right time. I’m skeptical that this is true. Their problems with depth extend back to last season. What seems more likely to me is that the Rangers are just in a tough spot in the schedule in terms of games played.