On November 14, 2015, Shayne Gostisbehere was recalled from the Lehigh Valley Phantoms and made his season-debut with the Flyers. The results of that call-up were nothing short of legendary -- a jumpstart of a moribund roster, 46 points in 64 games, and a second place finish in the Calder Trophy voting.
Almost a year later to the day, Gostisbehere found himself again in an unfamiliar home on gameday -- the Wells Fargo Center press box.
The news from Voorhees yesterday morning that Gostisbehere would be a healthy scratch against the Winnipeg Jets tonight understandably sent the fanbase into an uproar. Head coach Dave Hakstol's justification to the media present at practice was centered around "accountability within our group" and it being "a good thing for him in his growth and in his development and a few things we’re asking him to concentrate on." However, with 10 points in 17 games, it's tough to argue that Gostisbehere hasn't delivered positive contributions to the team thus far, at least on offense.
The prevailing theory is that the move is due to a perceived regression in the finer points of his game, particularly defensive play without the puck. Others have criticized Gostisbehere for missing too many shots from the point this season, and for supposedly looking a step slower than he did during his dynamic rookie year.
One of the best parts about hockey statistics is that we can test all of these theories. If Gostisbehere's play has regressed in any way, we should be able to pinpoint it somewhere in the numbers. And if no glaring weaknesses can be found, we can confidently disagree with decisions like the scratching of Ghost with true evidence supporting the position. So let's go through Gostisbehere's performance this season point-by-point, to see if we can find any concerning trends.
Have Gostisbehere's on-ice results gotten worse?
Let's start from the highest level possible. Raw point totals have never been the best way to evaluate a player, and that's even moreso the case now that we have a wealth of advanced metrics at our disposal.
Through the use of on-ice metrics, we can quantify how the Flyers have performed with Shayne Gostisbehere on the ice this season. While Gostisbehere certainly does not have a direct hand in every event that occurs on his watch, if good outcomes tend to happen when he plays, it's likely he's helping the team in a variety of ways.
Corsi For percentage, for example, evaluates how many shot attempts (blocked shots, unblocked shots, and shots on goal) that a player's team generates versus the total number of attempts created by both teams. So if the Flyers generate 10 attempts with Ghost on the ice and allow five, he will finish with a 66 percent Corsi For percentage on the night. Essentially, you at least want your players to finish above 50 percent, as that means that the team has been more productive from a shot creation standpoint than the opposition with said player on the ice.
Other statistics like Fenwick For percentage (same as Corsi, except it only includes shots on goal and missed shots) and Expected Goals For percentage (Fenwick but with each shot attempt weighted by the likelihood of it turning into a goal) serve the same purpose. Using these three metrics, we can determine how well the Flyers have carried play with Gostisbehere on the ice this year versus his fantastic rookie season.
As it turns out, Gostisbehere's on-ice metrics are actually better this year than last (all metrics are score and venue adjusted via Corsica.Hockey).
|Season||Gostisbehere Corsi For%||Gostisbehere Fenwick For%||Gostisbehere Expected Goals For%|
But raw on-ice metrics often don't tell the whole story. The Flyers as a team might be better in these three areas this season versus last year, and that's why Gostisbehere's percentage appear improved. To account for that, let's check Gostisbehere's relative metrics, which determine if the team's Corsi/Fenwick/xG has been better with Gostisbehere on the ice versus when he is on the bench. A high positive relative to teammates is good for Gostisbehere, while a negative would speak poorly to how Ghost has played.
Yet again, the Ghost Bear comes out better in 2016-17 than during his Calder Trophy season.
|Season||Corsi For% Relative||Fenwick For% Relative||Expected Goals For% Relative|
Last season, Gostisbehere was just barely negative relative to his teammates in all three key on-ice metrics. This season, he's positive in them all, a marked improvement. In addition, he's done this despite spending nearly 50% of his overall 5-on-5 ice time with Andrew MacDonald, who has been one of the league's worst drivers of play from the backend for the past three seasons. A quick look at the "With or Without You" tool on Stats.HockeyAnalysis.com shows that Gostisbehere's play-driving results have been dramatically improved away from Andrew MacDonald.
In 129 minutes with MacDonald, Gostisbehere's Corsi For percentage this year is 47.8%, but it leaps to a stellar 58.4 percent when paired with any other Flyers defenseman. And if you're more into using results like goals as a way to judge a pair, Philadelphia scored just 30 percent of the 5-on-5 goals while the MacDonald-Gostisbehere pairing was on the ice, but 57.1 percent of them when Gostisbehere was paired with anyone else.
What you're left with is a player who has driven play better this season than in 2015-16, in spite of being saddled with Andrew MacDonald for a month. Nothing yet to worry about.
Has Gostisbehere regressed defensively?
Metrics that take into account both offense and defense are the most valuable when it comes to fully evaluating a player. After all, a forward could be a liability in his own zone but make up for it with stellar shot creation on the attack, making him a net positive for the team. In the end, the goal of hockey is to outscore the other team, and whether that's by a 2-1 or 8-7 final score is ultimately irrelevant.
Still, it's fair to question whether Gostisbehere's defensive play has regressed this season, even if it's probably not the best way to judge him on the whole. If he's turning into a sieve defensively, however, it's understandable that the Flyers might want to address it now before the issue begins to turn him into an ineffective player, even if he hasn't quite reached that tipping point yet.
There are two ways to test this theory. First, let's take a quick glance at Corey Sznajder's manually-tracked individual player data for this season. Last year, Gostisbehere was the team's best defenseman in denying controlled entries into the offensive zone, and its best at moving the puck out of the defensive zone with possession. Both are key aspects of the "defensive" side of the position -- preventing opposing rushes, and getting the puck out of the most dangerous area of the ice. If Ghost is regressing defensively, maybe we'll see some sign of it here.
Again, not much to worry about here. He's dropped to fourth in the rankings when it comes to allowing controlled entries, but he leads the entire defense in Entry Denial percentage, so it looks like his aggressive instincts in the neutral zone are still there.
Next up, we'll test out his zone exit performance. Last year he was the Flyers' best defenseman in generating controlled exits, and was middle-of-the-road in avoiding turnovers or failed clears. Things haven't changed much here as well.
The only difference between this year and last is that now Gostisbehere has fellow phenom Ivan Provorov to contend with in the rankings. And when it comes to avoiding failed exits and defensive zone turnovers, he's actually third on the blueline in this category, trailing only Mark Streit and Brandon Manning. Yet again, we've come up empty in our attempt to find something wrong with Shayne Gostisbehere.
Let's pivot in our search for the truth. Rather than look at individual tracked metrics or all-encompassing on-ice metrics, we can analyze the latter but isolate for shot and chance suppression. Basically, we can check to see if the Flyers have given up more shots with Gostisbehere on the ice this season than they did last season. That could allow us to see if the 23-year defenseman is regressing from a defensive standpoint, which is maybe just getting overshadowed by larger gains made offensively.
We'll return to Corsi, Fenwick, and Expected Goals, but now we'll just focus on how many attempts the Flyers allowed with Gostisbehere playing, rather than looking at a for/against ratio. So has Ghost regressed from a shot suppression standpoint in 2016-17?
|Season||Corsi Against Per 60||Fenwick Against Per 60||Expected Goals Against Per 60|
Okay, now we've at least found something. While Gostisbehere has actually done a better job at suppressing shots from an overall volume standpoint, opponents have generated more Expected Goals with Gostisbehere on the ice this year versus last season. Still, we're not talking about a gigantic leap here, and he's made up for the increase in shot quality against by improving in on-ice shot quality for (2.79 this year versus 2.41 last season). But it is fair to say that opponents have generated a bit more quality offense on the whole with Ghost on the ice in 2016-17.
When you isolate the events that Gostisbehere is directly responsible for defensively (like defending zone entries and generating zone exits), he grades out fine. The numbers hint that his play in coverage hasn't been stellar, but they also show that those mistakes have been counterbalanced by strong shot creation, which Gostisbehere is almost certainly aiding.
What about his shots? Have teams "figured out" his point shot?
Another argument that has been thrown around recently (including by our own Bill Matz on BSH Radio this week) is the idea that Gostisbehere has been failing to hit the net far too often with his lethal shot, whether due to it getting blocked or Ghost just missing the target entirely. The theory is that teams are gameplanning to take away his shot and Gostisbehere is struggling to adjust.
Luckily, this is a theory that can be quickly tested by looking at Gostisbehere's individual shot results, both during 5-on-5 play and on 5-on-4 power plays. Has Ghost been missing the net more often in 2016-17?
|Season||5v5 Shot Attempts||5v5 Shots on Goal||5v5 Accuracy Rate||5v4 Shot Attempts||5v4 Shots on Goal||5v4 Accuracy Rate|
Hey, the eye test has a point! Gostisbehere's shot accuracy has been down a bit to start this season, both at even strength and on the power play.
But that leads to a bigger question -- is this dropoff anything to worry about? A 3.5 percentage point decline at 5v5 and a 4.5 point drop at 5v4 seems significant, but it's still pretty early in the season. Let's assume that just two more Gostisbehere shots at 5v5 and at 5v4 found their mark over the first 17 games.
|Season||5v5 Shot Attempts||5v5 Shots on Goal||5v5 Accuracy Rate||5v4 Shot Attempts||5v4 Shots on Goal||5v4 Accuracy Rate|
This should put things in perspective. We're not talking about a major issue for Gostisbehere, more a matter of a couple more shots here and there not making their way through the goaltender. And to dive even deeper into the numbers, while Gostisbehere has missed a few more shots this year than he did last season, he's actually generating better shots this time around.
|Season||5v5 Individual Expected Goals per 60||5v4 Individual Expected Goals per 60|
The moral of this story? Don't worry about Gostisbehere's shooting, either. Like in the other areas, he's absolutely fine.
After news broke that Shayne Gostisbehere would be a healthy scratch for the first time in his NHL career, the question had to be asked -- is there anything wrong with the dynamic young defenseman?
The short answer is that if there is an issue, it's not showing up in any of the stats. Gostisbehere has driven play at 5-on-5 better in 2016-17 than he did during his fantastic rookie season, and he's creating more dangerous chances for himself as well. That's even considering the extended time he spent alongside Andrew MacDonald, which unsurprisingly served to deflate all of his metrics. On the whole, the stats paint a picture of a player trending upwards, not one in desperate need of a dramatic adjustment to his game.
It does appear that opponents are generating higher quality chances with Gostisbehere on the ice this season versus last season, as shown by a slight jump in his Expected Goals against rate. But the Flyers are also creating more chances for themselves this year with Gostisbehere on the ice, as well. The result has been a net improvement in Ghost's Expected Goals ratio, mitigating the damage from any defensive issues.
In the end, that's a good description of Gostisbehere the player. While the hope is that he will improve defensively as he matures, it's unfair to expect that he'll ever be close to perfect in that area. Instead, the Flyers should be judging his play based on whether he helps to create enough meaningful offense to outweigh his deficiencies without the puck. The numbers support the claim that so far this season he's done just that, and actually seems to be improving his overall process from year-to-year.
Basically, healthy scratch or no, Shayne Gostisbehere is fine from a statistical standpoint. There's no real reason to worry that his game is regressing dramatically, or that he's dealing with a sophomore slump. Hopefully he can convince his coach of that fact in the coming weeks.