Neutral Zone Analysis: 2014-2015 Season: Game 5 @ Pittsburgh

Game 5 entries

Above are the Islander Neutral Zone results for Saturday’s game against Pittsburgh. The Isles basically played Pitt even in the neutral zone, or maybe were barely beaten as such while the game was close (Entries were 33-32 Pitt in close, 17-13 with control). As much as it felt like the Isles were losing at even strength, they didn’t really do so statistically by much. Just a failure of special teams, really. And this was against a top team in the league – Pitt is really good, much improved from last year.

Through 5 games, the Isles have basically beaten my expectations in ordinary fancystats. I mentioned that the goal for me was to get 6 points (they got 8) and to be above 50% fenwick close (they’re at 56.7%), so yeah that’s pretty nice. That said, their possession victories have come near entirely in the offensive zone – their neutral zone fenwick is just below even (49.8%), but they are massively outperforming what we’d expect from their entries in the offensive zone (the defensive zone #s are slightly below average).

So the goal for the next 5 games should be the following: Keep up above average possession numbers, 6 more points, and to win the neutral zone. The team should be now more used to how their linemates and d pairs play, and should get a healthy Lubo back soon, so this should definitely be attainable.


PS: I was asked in a comment to post who the above Islanders are in my neutral zone tracking. You can see the Islanders by Sweater # below:

Isles Roster #s

Neutral Zone Analysis: 2014-15 season: Game 4 Home vs San Jose

Game 4 entries

Game 4 Neutral Zone #s are below. The Isles outplayed the Sharks by a little bit in the neutral zone, but really outplayed them in the defensive and particularly the offensive zone, getting a ton of shot attempts off of each zone entry. Overall a great game in all 3 zones, with the Sharks’ scoring being pretty fluky. Well done vs a pretty good team.

PS: I was asked in a comment to post who the above Islanders are in my neutral zone tracking. You can see the Islanders by Sweater # below:

Isles Roster #s

Brock Nelson’s disappointing individual offense so far.

Brock Nelson has turned some heads to start this season – particularly those of the media who don’t know any non-Tavares Islanders. 4 goals in 3 games (now 4 in 4) demands some respect after all. But there’s a big number next to that 4 goal, 3 assist stat line that is a bit more alarming: Brock is shooting 80% in those 4 games.

Now anyone with 4 goals in 3-4 games is going to have a fluky high shooting % (unless your name is Steven Stamkos, apparently). That’s not the issue. The issue is this: if you’ve scored 4 goals, and have an 80% shooting %, that means you’ve only taken FIVE SHOTS ON GOAL. And two of Nelson’s goals have come basically on deflections (one in mid-air off his knee), so he arguably deserves to be credited with LESS shots!

This is a surprise. Coming into this season, there were two interesting questions when it came to shot totals. The first involved Mikhail Grabovski, whose shots and shot attempts per 60 tanked last year with the Caps. That question has been answered with a resounding positive so far: Grabovski leads the Isles with over 20 shot attempts per 60 at even strength (for comparison, the last 7 years he’s been at 13.4, and last year he was at 8.448).

The second involved Nelson, who was actually THIRD in shot attempts (corsi) per 60 last year for the Isles at 14.6 attempts per 60, but was only 7th with an average rate of SOG/60 since a high % of his shots were blocked last year. Research tends to support the idea that the % of shots of a player that are blocked is very random, meaning we would’ve expected Nelson to “take a leap” as more of his shots wound up on net. And it also seemed possible, with better linemates certain and his rookie year behind him, that Nelson would take an actual leap and increase his shot totals.

As I mentioned above, that hasn’t happened, and it’s a potential worry. Ignoring the power play (which should net Nelson extra shots, but hasn’t really yet), at even strength, only Josh Bailey, notorious for refusing to shoot, is taking less shots per 60 amongst Isles forwards – Nelson’s shot attempts per 60 through 4 games is only 10.16 – a 30% decrease from last year!

This is important because shot volume is the real key to goal scoring, not shooting %. Shooting %, as you should know, is incredibly fluky (leading to big hot and cold streaks), and your best shooters, barring the rare exception like Steven Stamkos, put up large #s by taking LOTS and LOTS of shots – see John Tavares for example. If Nelson is going to be a 20 goal scorer, year in and year out, he’d need probably to take around 200 shots on goal per year, or a rate of 2.43 shots per game. And that’s to be a TWENTY goal scorer, not exactly a high bar. Right now he’s achieving a rate at half that number, and that needs to change.

Now unlike a Josh Bailey, last year suggests pretty strongly that Nelson has the talent to do better than this. And Nelson’s line (especially when Grabbo comes back) will play well enough that even without scoring, he’s a good player. But by increasing his shooting, he could become truly great. And with Grabbo – a high volume shooter – now out, Nelson’s shot will become increasing important.

He needs to make use of it more. A lot more.

Neutral Zone Analysis: 2014-15 season: Game 3 @ NYR

Game 3 Entry data is below:


I don’t have much time to write, but I may come back to this game later. Ignoring all of the various negative and positive things that happened in the game, it was notable to see both Leddy (8 entries! 4 carry-ins!) and Boychuk (2 of those controlled entries are passes) being far more aggressive in this game. I like it a lot.

Unlike some of that game, such as the neutral zone D. Blech.

2 Game Samples are fun to talk about. They Ultimately mean nothing.

I had a tweet get retweeted by a lot of Isles people on twitter yesterday.

Unfortunately, I don’t think people got the point of those first two words – “For Fun”.  Corsi is a useful stat because it strips out things out of players’ control you have in goal based metrics (your goalie’s save percentage, for instance) and because the larger sample size makes it more predictive than goal based metrics at an earlier point.  That earlier point is NOWHERE NEAR 2 games – at around 8-10 games corsi starts to give us a reasonable picture to look at, and around 20 games is where we really start to have a picture that we can talk about somewhat confidently (this doesn’t mean things can’t change after 20 games of course!  But the sample is meaningful).

All the #s in that tweet say are what happened with those guys on the ice so far.  So while Leddy and Boychuk have been on the Ice, the Isles have so far been successful; with Strait and Reinhart, not so much.  That doesn’t mean that’ll continue.  I suspect it will, or I wouldn’t even have tweeted it, but we shouldn’t go overboard.

Calvin de Haan may be back before we get an 8 game sample of Strait-Reinhart (although I suspect Reinhart is the casualty here, due to his waiver exempt status, so we’ll see more of Strait).  But let’s not conclude just yet the pair is a big failure.

Neutral Zone Analysis: 2014-15 season: Game 2 vs Carolina

Game 2 entries

Above are the results of my neutral zone tracking for game 2 of this Islander season.  Again, for quick clarification, “controlled entries” are entries via carry-in or via pass – whereas uncontrolled are by dump or tip-in.  Dump-and-Changes are not counted.  The above is only 5 on 5 play.

Again, 1 game is not a sample size of any predictive value. But the point of these posts is to describe what happened during the game – when we get more data we can talk about prediction – how did the team and each player play in the neutral zone.  (This is not a complete picture there, which we’ll talk about in a future post.)

We have a larger sample of relevant data for this game, since two periods were played in “close situations” – the game was never out of “close” until Bailey’s goal at the end of the 2nd.  (For those who don’t know, “Close” is whenever the score is within 1 goal in the 1st two periods, or tied in the third.  In non-close situations, teams play differently, which can change how stats work).  At that point, the Islanders had basically played completely even with the Canes in the neutral zone at 5 vs 5, with the same amount of controlled entries (Carry ins or pass-ins) made by both teams, while the Isles had one more total entry.

You’ll note that’s not how the overall entries look above – this is a textbook example of “score effects” – given a lead in the third, the Islanders played less aggressively, being more prone to dump and more in a shell, leading them to lose the battle of both total entries and controlled entries.  This isn’t a flaw by the way, we’d expect the entries to be less effective of the Canes during this play by the Isles, and they in fact were – the Isles outshot the Canes in the third actually.  This is why for team data, we’ll be using “close” data.

Looking at the individual data, again there really wasn’t a dominant Islander on the night.  Grabovski led the way with 6 entries, but only 2 were carry-ins (a few were tip-ins).  Brian Strait by the way is credited with a carry-in which was actually a pass-in by goalie Chad Johnson – our tracking system has no way of crediting entries to goalies, so Strait is a lucky bum.

Speaking of which, it’s interesting through 2 games how little our D men are carrying the puck through the neutral zone.  Last Year Travis Hamonic carried in a bunch, and Nick Leddy was known for carrying in a LOT, while Thomas Hickey also was okay at carrying the puck in.  Leddy’s made one entry these two games.  I know it’s a riskier strategy to have your D carry in, and maybe it’s simply due to how the Canes play the neutral zone, but I’d like to see the D be a bit more aggressive in the future.  But we shall see.

Neutral Zone Analysis: 2014-15 season: Game 1 @ Carolina

Game 1 Entries

Above are the results of my neutral zone tracking for the first game of this Islander season.  For quick clarification, “controlled entries” are entries via carry-in or via pass – whereas uncontrolled are by dump or tip-in.  Dump-and-Changes are not counted.  The above is only 5 on 5 play.

The above #s are slightly skewed by score effects: the big lead made the team more willing to play and defend and dump the puck in – entries were 19-17 Isles, 7-5 by controlled entry while the score was still close, before the 2-0 lead.

Again, this is a small sample size so take it for what it’s worth.  Josh Bailey made the most entries, although he started the game with 3 straight dumps.  The new forwards managed 8 carry-ins themselves out of 14 total entries, which is a good rate.  Tavares is usually dominant in entries, but wasn’t last night – course he didn’t need to be.

Overall, a good game, just lacking relevant sample sizes.  We’ll talk more about this as we get more interesting data with larger samples.