During the opening weeks of the 2015-16 season, the biggest question surrounding the Philadelphia Flyers was the complete absence of goal scoring. Adjusting to a new coach and a new system, the team scored just 29 times in their first thirteen contests for a pitiful 2.23 goals per game rate.
This year, however, scoring has come far easier to Philadelphia.
Through the opening 13 games of the 2016-17 season, the Flyers have a whopping 45 goals, good for a stellar 3.46 per game rate. Their mediocre 6-6-1 record can be primarily laid on the shoulders of the defense and (especially) the goaltending, which has been gashed for 48 tallies against. Still, the offense has done its job.
The team has a whole has been an efficient scoring machine thus far. But which forwards have been most effective in the offensive zone, particularly during 5-on-5 situations? Answering this question allows us to better understand which players are standing out while on the attack, which players we can expect to continue to rack up points, and which ones need to make some adjustments.
The easiest way to determine who has been most effective in the offensive zone is simply to look at individual point totals. After all, if you're scoring (and assisting) on goals, you're probably doing something right while on the attack.
But there are some issues with using that as a sole approach. First, a player could be racking up lots of points simply because either he or his teammates are scoring goals at unsustainable high percentages. Sure, an on-ice shooting percentage of 20% is great in the here-and-now, but it's probably not going to continue over 82 games. Second, a player could be scoring not because he's especially effective in the offensive zone, but because he's getting there all the time due to strong neutral zone play and lots of faceoffs on that side of the ice. That still makes him valuable, but it's just a different type of value than a true offensive dynamo.
To isolate offensive zone performance from Flyers' forward during 5-on-5 situations, we'll use two methods of evaluation. To start, we'll determine which players have been most effective in helping the team to extract extra shots per offensive zone entry while on the ice. Then, we'll look at individual passing metrics to see which forwards are actively creating the most shots, whether by pulling the trigger themselves or passing the puck to a teammate who blasts away.
Offensive Zone Scores and what they mean
Last season, I manually tracked Flyers zone entries during 5-on-5 situations. Through that tracking, I was able to grade performance in every zone of the ice. To judge players' performance in the neutral zone, I counted every entry for and against that occurred with them on the ice. To measure their offensive zone play, I determined whether the Flyers generated more shots than would be expected considering the amount of times they made it into the attacking zone. And finally, I judged defensive zone play, evaluating if the team prevented more shots relative to entry-based expectations. Using past methods pioneered by Eric Tulsky, I was able to reproduce three metrics -- Neutral Zone Score, Offensive Zone Score, and Defensive Zone Score -- to quantify my findings.
This year, I am not tracking entries. That's because Corey Sznajder, microstat tracker extraordinaire, is doing the heavy lifting for everyone in the public sphere at his website, The Energy Line. However, though Corey's data includes basically every individual microstat imaginable, he does not record who is on the ice for each event, which is necessary in order to create metrics like Offensive Zone Score.
To create those metrics, Muneeb Alam of Japers' Rink has been kind enough to help match up Corey's Flyers-centric events to the on-ice data that the NHL tracks on their own. As a result, we now can determine which Philadelphia players were on the ice for each of the entries that Corey tracked, and recreate the Zone Score metrics used last season.
To evaluate offensive zone play, we'll obviously use Offensive Zone Score. Essentially, this metric determines how many unblocked shot attempts are expected to occur with a certain player on the ice, based on the number and type of offensive zone entries (including o-zone faceoffs) that happened on their watch, and then compares it to their actual results. If the team generated more shots than expected with him on the ice, then he gets a positive Offensive Zone Score. Less than expected, and he grades out negative.
Let's now look at each of the Flyers' forwards Expected Fenwick For (on-ice unblocked attempts), their actual Fenwick For, and then the resulting Offensive Zone Score.
|Forward||Expected Fenwick For||Actual Fenwick For||Offensive Zone Score|
There's a lot to digest here, so let's start with a high-level view. To begin, it's fairly impressive that 12 out of the 14 Philadelphia forwards are either positive or basically break-even in Offensive Zone Score. This implies that the team as a whole has done a solid job of extracting extra shots out of their entries into the attacking zone. The Flyers succeeded in this area last year as well, finishing with a +2.91% team OZS, and they're off to an even better start this year, coming in a +4.39 percent. Considering Philadelphia's aggressive forechecking ways, it's good to see that the tactics are succeeding.
As for the individual players, Roman Lyubimov obviously stands out. My eye test had him pegged as a particularly effective forechecker in the early going, and the numbers clearly back that up. Next up, we have the entire Konecny-Couturier-Voracek line, all of whom are solidly above-average in terms of offensive zone shot creation. This also should be no shock to even casual Flyers observers, as they've looked monstrous on the attack thus far.
Also interesting are the two players at the bottom of the list. Brayden Schenn was dropped off the first line recently, and his OZS shows that the Flyers have not been particularly effective in creating extra shots on the attack with him on the ice. Schenn does have the mentality and skillset to be a strong forechecker, but we haven't seen it manifest in the numbers so far this year.
Dale Weise's awful performance here also stands out. The offseason signing was a surprise scratch last week, but these metrics could help to explain why. Maybe Weise is struggling to adjust to the aggressive nature of the Flyers' offensive zone forecheck, and that's resulted in the winger moving into the coach's doghouse. Considering the ease at which NHL rookie Roman Lyubimov has seemingly helped his team to generate extra offense on the attack with a Weise-like skillset, it's surprising that Weise himself has graded out so poorly in this area.
The good news for Weise is that the jury is still out regarding the repeatability of Offensive Zone Score when it comes to individual players. It's clearly repeatable on a team level, but past research implies that individual players may not be able to consistently extract more value from their entries than their teammates do. My hope is that the current forechecking project that I am working on this season with Ryan Stimson and other writers at Hockey-Graphs will move this research area forward, but for now, OZS needs to be taken with a grain of salt. As a result, let's look to other methods of evaluating offensive zone play from the Flyers' forwards.
Introducing Primary Shot Contributions
One of the most enlightening projects in recent hockey analytics has been The Passing Project headed up by Ryan Stimson. Essentially a group tracking undertaking, The Passing Project is primarily focused on the impact of certain types of passes upon shot and goal creation.
One of the most intriguing findings so far has been the development of a new stat - Primary Shot Contributions per 60. The name is more threatening than the definition, which is extremely straightforward -- raw PSC is simply the total number of shot attempts and primary shot assists that an individual player creates. A "primary shot assist" is similar to a "primary assist," as it's the final pass that precedes a shot attempt by a teammate. PSC/60 (weighting the raw metric by time on ice) gives us an accurate view into which players are creating the most offense, whether by shooting the puck themselves or directly creating shots for teammates.
PSC/60 not only allows us to determine who is directly responsible for the creation of shot attempts for a team, it also has proven more predictive of future Primary Points for an individual player than even past primary points. Essentially, if Player A has a low Primary Points/60 rate through 10 games but a high PSC/60 rate, we can expect he'll score more points over the final 72 contests than a teammate who has racked up lots of primary assists with a low PSC/60. This makes intuitive sense -- if you're involved in the creation of lots of shots, over the long term, you're going to score points.
In addition to tracking entries and exits, Corey has also tracked shots and shot assists this season. Using that data, we can put together a ranking of Flyers forwards and their respective PSC/60 rates at 5-on-5, to determine who is actually creating the most offense and who can be expected to score a significant amount of primary points in the future.
|Forward||5v5 Primary Shot Contributions/60|
If you needed yet another reason, PSC/60 should get you really excited about the potential of Travis Konecny. No Flyers forward has created more shot attempts and primary shot assists at 5-on-5 this year, after accounting for ice time. Considering this and his stellar +11.87% Offensive Zone Scorer, it's not inaccurate to call him Philadelphia's best 5-on-5 shot creator so far in 2016-17, and this is as a 19-year old NHL rookie.
Aside from Konecny, the top of the chart reflects Philadelphia's "top-six" forwards so far this year. Voracek, Simmonds, Giroux and Couturier are no shockers here, but the appearance of Michael Raffl over Brayden Schenn is a bit of a surprise. As it turns out, the numbers backed up Dave Hakstol's decision to bump Schenn off the top two lines in favor of Raffl, who has done a much better job of directly creating tangible offense so far this year.
Matt Read's hot offensive start appears to be a bit of smoke and mirrors, as he still is grading out as a bottom-six forward in terms of offensive creation, even if he's been his usual useful self in the middle of the ice in terms of driving play. But yet again, the eyesore here is the performance of Weise, who has been less effective in creating shots at 5-on-5 than even Boyd Gordon, he of the four points in 65 games last season. The Flyers need to hope that this has just been a slow start for Weise, because he's under contract for three more years after this one.
PSC is predictive, but does it isolate offensive zone play?
The best attribute of PSC/60 is its ability to predict future primary points on an individual level better than past primary scoring itself. However, for the purposes of this exercise, it does not perfectly isolate offensive zone performance. After all, a player could be racking up lots of PSC tallies simply because he's getting into the offensive zone more than his teammates, not because he's been especially efficient once there. In that case, his strong PSC/60 would be more a product of stellar neutral zone play than effectiveness on the attack.
By the same token, a player could have a PSC/60 that looks mediocre, but he's actually quite useful once in the offensive zone. He maybe has just struggled to get there in the first place. This could be the type of player that might be best served receiving a heavy dose of faceoffs in the attacking zone, in order to put him right in his comfort zone.
But how do we strip away neutral zone (and zone start) impacts from PSC? The easiest way would be to divide a player's total number of PSC events from the amount of times he has been on the ice for an offensive zone possession. For example, Sean Couturier has been on the ice for 268 Flyers offensive zone entries and 58 offensive zone starts. Since he has generated 69 primary shot contribution events, that means he's averaged 0.21 PSC events per entry.
Even this is an imperfect method, though, because not all entry types are created equal. We know that controlled zone entries generate an average of 0.66 unblocked shot attempts, while uncontrolled entries (dump-ins) and faceoffs generate about 0.29 unblocked attempts. As a result, let's weight each entry in our formula by those rates to truly isolate offensive zone PSC creation.
Konecny remains by far the most impressive Flyers forward, even after correcting for number and type of entries. But our adjusted list gets interesting soon after. Pierre-Edouard Bellemare, uninspiring in PSC/60, jumps all the way to second in our new measurement, above Voracek, Simmonds and Giroux. Regular linemate Chris VandeVelde also moves into more than respectable territory, in fifth.
At first glance, this is a shock. But PSC has liked Bellemare in the past. The work done by Alan Wells this past summer on Stimson's data showed very clearly that the French forward graded out in the 84th percentile in PSC/60 last year, much higher than most would have expected. We're not dealing with incredibly large samples here, but it's looking legitimately possible than Bellemare may have some untapped offensive zone potential, at least in terms of shot creation.
Moving further down the list, Sean Couturier drops out of the high-end tier and more into the muddled middle once we control for his always-stellar neutral zone play. He's not a useless forward in the offensive zone by any means, but he certainly could stand to be a bit more assertive there. If that's not possible, he'll have to continue to use his neutral zone dominance to make up for his relative weakness in Isolated PSC.
The bottom of the list is enlightening for what it may tell us about Dave Hakstol's evaluation preferences. The bottom five players on the list are Gordon, Read, Cousins, Schenn and Weise. Out of those five, three have been scratched at least once by Hakstol, one has been demoted from the top-six, and Read is riding a sure-to-decline 23.1% shooting percentage to offensive usefulness. It's just a theory, but this metric may be giving us a glimpse into what is driving the Philadelphia coach's decision-making at forward. The more active you are in the offensive zone at 5-on-5, the more job security you seem to have.
When evaluating the best Flyers' forwards in the offensive zone so far this season, the conversation begins with rookie Travis Konecny. With Konecny on the ice, Philadelphia has averaged 11.87% more unblocked shot attempts than would be expected, and no Flyers player has directly contributed to the creation of more 5-on-5 shots than Konecny thus far. The stats back up the eye test here -- every time Travis Konecny hits the offensive zone, he's been a fright for opposing defenses.
Right behind are the usual suspects. Jakub Voracek has been Konecny's partner-in-crime, ranking second in PSC/60 and third in isolated PSC, while holding a +8.80% Offensive Zone Score. Wayne Simmonds, Claude Giroux and Michael Raffl also have been effective in the offensive zone, while Sean Couturier has been more of a support forward while on the attack, helping his linemates to get into the opponents' end and then letting Konecny and Voracek take the lead on creating shots.
Pierre-Edouard Bellemare and Roman Lyubimov are the most intriguing forwards. Bellemare's Offensive Zone Score and PSC/60 are nothing special, but his Isolated PSC trails only Konecny. Considering his strong performance in PSC last season, Bellemare's skillset is getting harder and harder to pin down. It seems like he may be above-average at helping to generate shots in the offensive zone, but his career 0.72 Points/60 at 5-on-5 implies that very few of those shots end up in the net.
As for Lyubimov, he hasn't factored into the creation of many shots, but his +30.99% Offensive Zone Score leads all Flyers. He could be due for a big regression in terms of play-driving, or maybe he is so good at forechecking that he can help his linemates to generate extra shots without directly creating them himself.
The three most disappointing players so far have been Brayden Schenn, Nick Cousins, and Dale Weise. Schenn was a disaster to start the season, posting just 10 primary shot contributions in his first four games. His offensive zone play has picked up recently, and the Flyers will be hoping that positive trend continues, especially because of the substantial financial commitment they made to Schenn in the offseason. Cousins' offensive zone play has been comparable to that of Boyd Gordon, which is certainly not company that the 23-year old wants to keep from a shot creation standpoint. He does have the excuse that he's spent lots of time at wing (a new position for him) but his offensive game simply needs to be better on the whole.
Weise has been the biggest issue. Not only have the Flyers struggled to create shot attempts with Weise on the ice (-17.82% Offensive Zone Score), Weise himself has barely factored into the offense. He's last on the Flyers so far with a PSC/60 of 15.12, and grades out even worse in Isolated PSC. Interestingly enough, Weise's play in the neutral zone has actually been okay (54.84% Controlled Entry Rate), but he's giving back all of those gains due to utter invisibility in the offensive zone. Keep an eye on Weise the rest of the way to see if his shot creation game continues to be nonexistent, or if he begins to assert himself more in the coming weeks.
All derived metrics are courtesy of Corey Sznajder, Muneeb Alam, Corsica.Hockey, and NaturalStatTrick.