Islanders Regular Season Review via Analytics Part 1: Neutral Zone Play (Now with Neutral Zone +/-)

"Am I not a joy to play with?" - Bruce Bennett

Well guys, the season is over and I'm going to be starting a series of posts reviewing the Islanders' individual performances through Hockey Analytics - what some might call "Advanced Stats." I'm going to start these posts with a review of the Islanders' Neutral Zone Play, which I have numbers for thanks to the zone entry tracking I've done for the team this year.

If you don't know what zone entries are, I'd encourage you to read my intro post about them, but if you'd rather not for some reason, I'll explain quickly here:

Zone Entries are the name given to each entry made by each team into the offensive zone from the neutral zone. In effect, I'm going through each game and tracking each time the puck travels from the neutral zone to the offensive/defensive zone.

What do I mean by tracking? Well, ,what I mean is that I'm tracking who gets the puck over the blue line, how they do so (via dump, tip, carry-in, or pass), and whether it's even strength or not.

The whole point of this exercise is that it essentially gives us a method to measure which players are winning the battle of the neutral zone, which is incredibly important to the game of hockey, but basically unmeasurable by traditional statistics. Teams that win the neutral zone win more games because they get more time in the opponents' zone and manage to get more chances to score than their opponents.

How do we tell if a team is winning the neutral zone? Well, quite simply, the better neutral zone teams not only get the puck more often into the opponents' zone, but they also get it into the opponents' zone with POSSESSION. In other words, better teams will carry or pass the puck into the offensive zone more often than they dump the puck in. Getting the puck into the zone with possession results in more than double the amount of shots on goal than getting the puck in via dump-in (or tip-in), so it's a major factor in winning hockey games.

Now you might recall that I did a quick post on Isles Neutral Zone Play right before the playoffs? So why am I starting with this again?

The reason is that the picture I have painted in previous posts has been incomplete - I have had the Isles individual offensive neutral zone play #s - the zone entry results for each Islander - but I have NOT been able to present the defensive side of the Neutral Zone. In other words, it's great to know that John Tavares carries the puck in a ton of the time - most on the team - and makes a ton of entries - most on the team - but all that may not be that effective if he allows the opponents to do the same!

Fortunately, thanks to Muneeb Alam of Red Line Station, I now have the total neutral zone #s. So let's go through them.

First, let me present a graph showing something I've shown before:


What does this graph show? Well on the Horizontal Axis, we have the % of entries while a player is on the ice that are made by each player. In other words, the further right on the graph, the bigger burden the player has in bringing the puck through the neutral zone into the offensive zone (by pass, carry, tip or dump). You'll note all the D-Men are on the left side of the chart, as naturally most of this job is taken up by forwards.

On the vertical axis we have the % of entries by each player that were made with possession. What does this mean? Well we mean the % of entries into the offensive zone by each player that were made by carry-in or pass-in, rather than by tip-in or dump-in. In case you wonder why this is a big deal, here were the #s for the Islanders over these last 48 games:

Entry Type
Shots Per Entry
Goals Per Entry
With Possession
0.57 0.040
Without Possession
0.23 0.008
0.25 0.013

As you can see, when the puck enters the zone while the Isles have possession, the Isles manage nearly 2.5x the amount of shots and 5x as many goals as when they dumped the puck in.* That's kind of significant.

*The goals # is a bit fluky - Eric T showed the # more likely to be around .013 and my own #s for Isles opponents is .013. So really it should be an Entry with possession obtains three times as many goals as a dump-in - which is still pretty damn big.

So getting back to our graph, you can see here the roles of each of the Islander players and how successful each player was at the role offensively. Tavares took on the heaviest role of any Islander and still entered the zone with possession at the highest rate of any Islander. That's impressive. By contrast, his frequent linemate, Matt Moulson, takes a TINY role in getting the puck into the offensive zone - the least of any Islander forward - and doesn't perform particularly well - dumping more often than not. We'll get to the actual #s below, the graph speaks for itself.


Second, let me show you a graph that I couldn't have shown you without the data I recently obtained - showcasing the defensive part of the Isles' Neutral Zone Play:


This graph showcases the Neutral Zone play of OPPONENTS while each Islander is on the ice. The Horizontal Axis measures how many entries into the Isles' zone are made by opponents (per 60 minutes) while each player is on the ice. The further right, the more often the opponent is getting the puck into the Isles' own zone.

The Vertical axis again measures the % of these opponents' entries that are made with possession as opposed to via dump. The higher the #, the more often the opponent gets the puck into the Isles' D zone via more dangerous carries and passes rather than via dumps.

Because this graph represents defense, the best area of the graph to be in is the bottom left side of the graph (as opposed to the top right like the first chart). You'll notice a trio of players in that sort-of-vicinity: Kyle Okposo, whose presence on the ice limited opponents' entries more than any other forward (although opponents still managed to enter with possession 50% of the time), Lubomir Visnovsky, who allowed the least amount of opposing entries and also suppressed entries with possession, and Thomas Hickey, who wasn't as good as Lubo in preventing entries (but was still pretty damn good) but prevented opponents from carrying it in the most of any Islanders' player.

By contrast, the right and top right parts of the graph are problems. In the far right you see goon Eric Boulton, who can't stop opponents from skating right by him worth a damn (although opponents did dump it in a lot vs Boulton, the sheer amount of extra entries they got against him makes this pretty awful). In the top right, however you see the Isles' "Shutdown D-Pair" AMac and Hamonic. Yes the two players played against top competition, but even so, they were clearly badly overwhelmed by it in the neutral zone, allowing a lot of entries and a ton of them to be with possession. Not what you want out of your "Shutdown Pair".


Third, let me show you a graph putting the above two ideas together, showcasing both the offensive and defensive neutral zone play of each Islander player:


On the horizontal axis of this graph we have the label "Defensive Neutral Zone Fenwick." What the hell does that mean? Well it represents the amount of unblocked shots (Fenwick) aimed at the Isles' net that we would expect from the neutral zone play of each player while they were on the ice. In short, the farther right you are on the graph, the worse you are at defensive neutral zone play. The farther left the better.

The vertical axis is the offensive version of the same thing. The higher the value, the more unblocked shots we'd expect by the Isles at the opponents' net based on a player's neutral zone play. The higher the better the team does in the offensive element of neutral zone play. The lower the worse.

The Blue diagonal line btw represents the break even point. If a player is below that line on the graph, the team is losing the neutral zone battle with him on the ice. If a player is above that line, they're winning it. You'll notice the furthest below the line is goon Eric Boulton. However, big guy Joe Finley isn't far off in awfulness. The same is true of both Matt Martin and Matt Carkner, two other "big guys" - apparently being big isn't necessarily a plus in the neutral zone - Martin, Carkner, and Finley aren't terrible defensively in the neutral zone, but they can't get the puck into the opponent's zone effectively worth a damn. Incidentally you'll notice the Isles' Shutdown pair of Hamonic-AMac again below the blue line, due to their lousy defensive numbers discussed above.

By contrast, the furthest above the line belongs to a few already mentioned players - Visnovsky & Hickey as D Men, Okposo at forward (with linemates Nielsen and Bailey significantly above the break-even line as well).


A good question by the way asked in the comments by Afrosupreme recently should be discussed here: "How do zone starts affect these #s?" For those who aren't familiar - "zone start %" is the % of shifts each player starts in an offensive zone faceoff as opposed to a defensive zone faceoff. Coaches regularly get their best offensive players - like the Tavares line - more offensive zone faceoffs while their better D Zone players get more Defensive ones. Tavares for example, had nearly twice as many offensive zone faceoffs as defensive (61.2%). This of course leads to more shots and goals.

Oddly, the more offensive zone faceoffs you take, the lower your Neutral Zone #s - the reason for this is that if you start a shift on an offensive zone faceoff, when the puck goes back to the neutral zone it's now going the other direction and is likely heading into your own zone. Thus one might figure a player like Tavares would have worse Neutral Zone #s above on that graph (and in the #s below) because of his usage.

However, despite the last two paragraphs, it appears that the effect is minor.* Tavares had the most extreme offensive zone starts of any Islander but even his Neutral zone #s barely changed (His Neutral Zone Fenwick, which I'll explain below, went from 50.3% to 50.7%). As a result, I'm not going to bother talking about the zone-start #s below when I give the #s for each individual player.

*For those who care, the effect of an additional offensive zone faceoff is to reduce neutral zone fenwick by .09 shots. And this # is in fact overstating the effect - because the longer a shift is of a player, the less effect of the offensive zone faceoff. So the effect has little impact on the results at .09 shots per faceoff and the effect is probably lower than that....well, it's not worth talking about.


Okay enough of the graphs, let's go into the actual numbers of the players. If the above was a bit much to digest by the way, go ahead and take a break and come back when you're ready. This is a good place to break.

NAME Individual Burden % of Individual Entries with Possession On-Ice Entry % % of On-Ice Team Entries with Possession % of Opponent's Entries with Possession Neutral Zone Fenwick %
John Tavares
32.6% 73% 50.4% 55.2% 55.0% 50.4%
Matt Moulson
19.6% 38% 50.5% 54.6% 54.8% 50.4%
Brad Boyes
28.2% 53% 49.8% 54.1% 52.9% 50.0%

What do all these #s mean? Let's go through each one:
Individual Burden: % of a team's entries made by that player - see the first graph above.
% of Individual Entries with Possession: % of entries by that player made by carry-in or pass-in.
On-Ice Entry%: The % of entries while a player is on the ice that go into the offensive zone. Above 50% means that more entries are made into the offensive zone than defensive zone, below 50% means the opposite.
% of On-Ice Team Entries with Possession: % of Isles entries while a player is on the ice that were made with possession - by carry-in or pass-in
% of On-Ice Opponent's Entries with Possession: % of Opponent's entries while a player is on the ice that were made with possession - by carry-in or pass-in.
Neutral Zone Fenwick*: A measure that takes the previous three measures and uses the average results of each type of zone-entry to determine whether the team is winning or losing the neutral zone battle. 50% = even, above 50% = the Isles are winning with a guy on the ice, below 50% means the Isles are losing the NZ battle with the guy on the ice.

*Neutral Zone Fenwick % essentially measures the expected possession #s of each player based upon the Neutral Zone +/- #s.


Enough with definitions, how did the first line do? Well, the first line's results are interesting. As noted above, they get fractionally better if you factor in zone-starts, but just barely. Notice, despite Tavares' dominance with the puck in the neutral zone, the line is barely winning the neutral zone battle at all! How is that possible?

Well, first of all, Matt Moulson takes such a little role on this line - and does it poorly, that I suspect that it makes the Neutral zone battle harder for everyone else. This isn't saying Moulson isn't a great player, but this is not his strength.

But really the big issue here is that while this line is getting the puck into the opponent's zone with possession at a great rate, Opponents are doing the same in return and are making a lot of entries themselves. It does seem, at least in the neutral zone, that it's true when people say that this line is strong on O but weak on D. Overall though, the O outweighs the negatives of the D, and when you combine that with scorers/passers like Moulson/Tavares, you get good things.

NAME Individual Burden % of Individual Entries with Possession On-Ice Entry % % of On-Ice Team Entries with Possession % of Opponent's Entries with Possession Neutral Zone Fenwick %
Frans Nielsen
25.5% 68% 51.2% 53.9% 49.1% 52.0%
Kyle Okposo
29.8% 60% 51.9% 55.3% 49.9% 52.8%
Josh Bailey
24.8% 56% 50.8% 54.6% 48.7% 51.9%

Now this is more like it. Neutral Zone Fenwick #s 51% and above are good, #s 52 and above are great/elite. And these guys were playing against top opposition forwards too.

Notably, unlike the first line, the neutral zone roles on this line were far more balanced - Okposo took a greater role than the other two guys, but not by a crazy amount and no one was Moulson level. Meanwhile, all three guys were strong at getting the puck in with possession. Not only that but they prevented the opponents from doing the same, and probably not coincidentally, out entried the opponents.

NAME Individual Burden % of Individual Entries with Possession On-Ice Entry % % of On-Ice Team Entries with Possession % of Opponent's Entries with Possession Neutral Zone Fenwick %
Keith Aucoin
25.8% 46% 50.2% 48.3% 50.0% 49.8%
Michael Grabner
29.1% 55% 50.2% 50.7% 50.8% 50.2%
Colin McDonald
30.4% 40% 50.3% 44.3% 51.2% 49.0%

This was the third line for most of the final stretch of the season and the playoffs, although all three guys played significant time with other teammates, hence why the #s aren't as tied together as the previous two lines. Both Grabner and CMac took large roles in the neutral zone, but Grabner's speed allowed him to carry it in a lot more while CMac mainly used dumps. The end result is that CMac lost the neutral zone battle by a little bit, while Grabner was basically even. Aucoin, fwiw, was basically even as well, although he took a lesser role than the other two guys.

Overall, this unit will probably be re-shuffled next year (due to Aucoin's likely leaving and the appearance of new players) and there's nothing here that's particularly worth keeping (Nothing's poor either, mind you).

NAME Individual Burden % of Individual Entries with Possession On-Ice Entry % % of On-Ice Team Entries with Possession % of Opponent's Entries with Possession Neutral Zone Fenwick %
Casey Cizikas
27.0% 46% 48.9% 43.1% 45.6% 48.4%
Matt Martin
25.1% 31% 48.7% 39.5% 44.1% 47.8%
Marty Reasoner
20.8% 49% 48.6% 44.5% 45.9% 48.3%

The fourth line #s (ignore Reasoner here) are particularly interesting. They've lost the neutral zone battle by a good bit, especially Matt Martin whose 47.8% is the worst by a regular Islander forward. As you might expect this is pretty much entirely on the offensive end - opponents are finding it really hard to carry in against this line, but this line can't seem to carry it in the other way.

Now I've seen suggestions that Martin and his line's dumping in isn't a problem, because those guys aren't trying to score anyway. But here's the problem: even if you think the dumps aren't an issue (and they are - a 39.5% entry with possession rate is horrible), the problem is that this line isn't MAKING enough entries overall! In other words, even if this line's role could be described as get the puck in the offensive zone and stall there till you line-change, the opponents are getting more entries than Martin and his line overall with him on the ice by a decent amount! So they can't exactly be doing their jobs very well. (More on Martin in later posts in this series)

NOTE: These type of #s suggest this type of line is really only useful for using in the defensive shell game late with a lead. In those situations, you don't mind the lack of O as much when this line can reduce the amount of dangerous opponent entries that are likely to lead to an equalizing goals. But if you're behind late, this line should be shelved as much as possible. Capuano didn't ever do that.

NAME Individual Burden % of Individual Entries with Possession On-Ice Entry % % of On-Ice Team Entries with Possession % of Opponent's Entries with Possession Neutral Zone Fenwick %
Jesse Joensuu
25.6% 68% 46.9% 50.0% 40.2% 48.7%
Anders Lee
25.0% 17% 52.0% 46.2% 50.0% 51.3%
Eric Boulton
25.7% 29% 47.1% 33.9% 41.9% 45.5%
David Ullstrom
32.4% 59% 48.9% 50.2% 47.2% 49.4%

These four guys are all guys who played only small bits this season and thus have small sample sizes, so take these with a heavy grain of salt (Anders Lee's #s are basically nothing). Ullstrom's #s are by far the closest to being significant here - and they're pretty solid - just losing the neutral zone battle overall, but mainly due to total # of entries against rather than due to % of entries with possession. And he takes a heavy neutral zone burden - just like John Tavares. He could be a nice depth piece for a team someday - if he gets playing time.

Joensuu's #s are also potentially nice, but he really lost the NZ battle due to the sheer amount of opponent entries. FWIW this may have been the result of his frequent linemates (See above): Casey Cizikas and especially Matt Martin.

Boulton is just....Boulton. Enough Said.

NAME Individual Burden % of Individual Entries with Possession On-Ice Entry % % of On-Ice Team Entries with Possession % of Opponent's Entries with Possession Neutral Zone Fenwick %
Travis Hamonic
13.7% 35% 49.5% 51.8% 55.7% 49.0%
Andrew Macdonald
8.3% 23% 49.9% 50.9% 55.2% 49.2%
Thomas Hickey
8.6% 47% 51.3% 52.9% 43.5% 53.1%
Lubomir Visnovsky
13.6% 57% 51.5% 52.3% 45.6% 52.8%
Mark Streit
13.7% 57% 49.5% 51.1% 49.5% 49.8%
Brian Strait
8.3% 30% 50.9% 50.4% 47.0% 51.5%
Matt Carkner
10.7% 17% 48.1% 44.8% 48.3% 47.4%
Radek Martinek
6.2% 29% 50.2% 51.4% 48.3% 50.7%
Joe Finley
6.8% 6% 47.9% 41.9% 45.9% 47.1%

The Defensemen are....interesting. Each D pair consisted of a primary puck-mover: Visnovsky, Streit, and Hamonic, who each had near identical neutral zone burdens, and one secondary puck-mover taking a much lighter role. Note that Carkner seemed at times to pull a middle ground, becoming the primary puck mover with certain other D-Men.

As noted above, AMac and Hamonic played mediocre in the neutral zone. Yes they faced tough opponents - mind you, Nielsen's line played tougher and succeeded) but opponents were having their way with the shutdown pair in the D Zone: entering with possession a ton (over 55%!) and out-entering the Isles slightly. These are concerning #s - and another sign that things went wrong for this pair after success in prior years.

By contrast, Lubo-Hickey was a dominant pair, with two of the best neutral zone #s of the entire team (Hickey's 53.1% NZ Fenwick is a team high, Lubo's 52.8% is tied only with KO). Both players were good puck movers, infrequently resorting to dumps (a great feat for defensemen) and both guys HEAVILY suppressed opponent entries: with them on the ice the Isles managed over 51% of entries, and most of the opponent's entries were of the much less dangerous dump-in variety.

Mark Streit is....interesting. His individual #s are identical to Lubo's. His defensive #s aren't bad. So why is he below the break-even point? The problem appears to be his d-partners. Here's Streit's NZ Fenwick with various partners:
Carkner: 46.6%(17.4% of Streit entries)
Finley: 48.2%(14.1% of Streit entries)
Martinek: 49.9%(24% of Streit Entries)
Strait: 52.6%(23.1% of Streit Entries)

With Strait, Streit's NZ Performance - in a tiny sample mind you - was near Lubo-Hickey range. With Finley and Carkner it was abysmal (with Martinek it treaded water). Streit is still obviously a useful piece for neutral zone play - although he requires a not awful D partner - he just was saddled with awful baggage. Brian Strait however should be a joy for some D partner next year to play with.

Speaking of awful baggage, some have suggested that Matt Carkner's bad neutral zone #s were due mainly to him letting better puck handlers control the puck and thus it was okay. The problem is that he was sometimes the primary puck-handling D Man, and well overall the team struggled to make offensive entries in the neutral zone with him on the ice. Not a good sign for a guy with 2 more years.

Oh and the less said about Joe Finley, the better.


Overall, the Isles managed to outplay their opponents in the Neutral Zone, with a 50.7% Neutral Zone Fenwick overall. The top Neutral Zone line should return unchanged next year, but there are changes coming (Streit for instance) that could change things a bit.

<em>Submitted FanPosts do not necessarily reflect the views of this blog or SB Nation. If you're reading this statement, you pass the fine print legalese test. Four stars for you.</em>