Welcome to the third part of my mid-season review of the Islanders: Forwards. The Isles' forward corps seems to be unproductive outside of John Tavares. But is that really the case? After all, this team is barely below even in terms of possession and is still scoring at a good clip at EV. Where are the problems? Who's not holding his own?

Let's go through each forward one at a time.

Note: The following stats are from Behindthenet.ca except for the zone entry #s, which are from my own tracking. A caution: the Zone-Entry #s are 3 games behind (in part because of the time it's taken to write this post).

Also: If the tables below are cut off horizontally, try pulling your browser window wider -- the site's responsive design should let you see them.

So let's start with the top:

Tables 1 and 2: Relevant Stats and Fancy Stats of John Tavares.
LEGEND:
Corsi/60: Shots +/- using ALL Shots - so the statistic is (Blocked+Missed+On Goal Shots) For - (Blocked+Missed+On Goal Shots) Against per 60 while a player is on the ice.
Relative Corsi/60: A Player's Corsi/60 MINUS the Team's Corsi/60 while he's not on the ice.
Zone-Start %: The percentage of faceoffs you're on the ice that are in the offensive zone (not including neutral zone faceoffs)
Quality of Competition (Relative Corsi): A measure of the skill of the average opposing player on the ice. (Basically it's the average Relative Corsi of the opposing players on the ice while this player was on the ice.)

So last year Neil Greenberg used some possession numbers as part of his case against Tavares' inclusion in the top 25 under 25. As I pointed out then, that was kind of ridiculous - while JT 's Corsi on was basically even, it was around #1 on the team with weak players behind him - especially on defense. Yes, he was a little "sheltered" but his relative numbers showed his production was beyond what we'd expect from those minutes given to Average player. And notice I've yet to even mention the fact he was near point per game at the time.

This year no context is really needed to see Tavares's superstardom (but I'll give some anyhow). Tavares is getting slightly more sheltered minutes than last year- with nearly 60 percent of his face offs coming in the offensive zone. Of course again this is nothing compared to the advantageous minutes seen by guys like Rick Nash or the Sedins (70%)

Of course what he's doing with those minutes is ridiculous. The isles are outshooting opponents by 7 shots per 60 while Tavares is on the ice, and is nearly 15 shots per 60 better with JT out there than without him at 5 on 5. Yeah that's really good. Like really really good.

He's also drawing a ton of penalties 11 compared to 5 taken at 5 on 5 which essentially creates another isles goal (considering roughly 1 goal is scored per every 5 power plays)

Oh, and did I mention Tavares is also a master at neutral zone play? Leads the team both in entries and entries with possession (71% of entries are by carry or pass rather than by dump).

And this is before we talk about JT's 16 goals and 13 assists. Which is pretty damn good though you don't need me to tell you about that.

Two cautions must be talked about when talking about Tavares however.

First, a fake caution- don't be misled by JT's plus/minus - that's entirely the product of the team's goaltending.

Second, a real caution: Tavares can't keep up this goal scoring pace. Right now JT has a 16 percent shooting percentage. His career average is around 12%. Expect the latter going forward- as dangerous a shooter as he is he hasn't improved that much that's a ginormous improvement which generally doesn't happen in the nhl.

This doesn't mean jt won't score more goals per game this year and the next - in addition to having a fluky high shooting percentage, JT has added about an extra half shot per game on goal to his game. At his career shooting rate that's about 4-5 more goals, a big deal. What a stud.

-----------------------

Matt Moulson naturally plays nearly all of his minutes with JT. As such it shouldn't surprise anyone to see Moulson's possession numbers are near identical to JT's (marginally worse for obvious reasons).

Unlike JT, Matty Mo does not excel outside of the offensive zone. Moulson isn't completely out of his depth but when the puck is on his stick in the neutral zone he prefers to dump the puck in instead of carry. But with JT on his line this isn't that big a deal as his line mates carry the puck in far more frequently.

Oh and unlike JT, Moulson's #s are only going to regress UPWARDS: His shooting % is down this year and will go up to his normal 14% level as the season continues, and did I mention he's drastically increased his ability to get his shots on net? If he could maintain this shot on goal pace while his shooting percentage goes back to normal, Moulson would be on pace for 41 goals. 41!

Brad Boyes was an interesting signing on the offseason and I think a successful one (whether he's on the team next year is another question). His numbers above are really misleading and should be split into two time periods:

Games 1-8 were when Boyes was paired with Grabner and Nielsen. This was a more defensive role than you'd expect from Boyes and well, the line bombed badly -- together with Frans the team had a zone adjusted Corsi % of 35.2%. What that means is that while that line was on the ice, roughly 65 percent of shots taken on the ice were at the isles' own net. Not good. It didn't help that Boyes was acting like Moulson in the neutral zone and dumping the puck inmost of the time.

From Game 9 onwards, Boyes replaced KO as the 2nd winger on the JT line and he has been much better. It's obviously hard to separate these numbers from JT and MM's but the two have been better with Boyes than with Okposo. Moreover, Boyes has increased his amount of entries by carry and pass and decreased his dumps.

It goes without saying that Kyle Okposo has been the most controversial Islander this year. And the #s do bear that out somewhat. The JT line was performing worse with Okposo than Boyes (it had a negative possession rate with Okposo despite the cushy minutes), which is not the best of signs. Okposo's entry with possession rate is down from the last 10 games of last season, although it's still quite solid. But again, below expectations. And this is without going into the scoring.

Mind you, since moving to the Frans Nielsen line, Okposo's #s have gotten better such that the #s above don't tell the whole story. KO's possession #s, when adjusted for zone-starts, are basically even (just below actually) with Frans, whereas with JT they were clearly below even. But again, these minutes with Frans have been in oddly offensive minutes, where you'd EXPECT better possession #s.

In short, the #s above are fine. But the team itself has improved around Okposo and you'd like to see his #s improve along with them which they haven't. And at age 25, KO is hitting the age where Hockey players PEAK and stop getting better. Uh oh.

Now I'm less negative about KO than others are for a few reasons:

1. He's not a 3.9% shooter - his shooting percentage will regress upwards and shots will find the twine. Goals will come at a higher rate than this. On the other hand, even if KO had his average Shooting % this year, he'd be at 4-5 goals right now, on pace for less than 20 in a season. That's not what you'd hope for.

2. This is a 26 game sample size. That's decent, but it's still not enough that you throw out the #s and results from years prior. Players can have bad streaks even over such a time period and they snap out of it when that happens. In short, KO's possession #s probably should be expected to improve a little bit as they were kind of better - (compared to his linemates) - last year. Of course, he wasn't that much better, so I'm not sure it'll be that much of an improvement.

In short, yes KO has been a disappointment. But at this point the Isles seem best served playing him more and more, hoping he gets on one of his hot streaks as positive regression kicks in, and THEN seeing if they can deal him. His value is at its lowest right now, there's no point in dealing him.

The other Islander enigma, Josh Bailey, is yet again confounding expectations. There's both some good and some bad here, so let's start with the good - which oddly enough lies in Bailey's offense. Yes, 5 points in 15 games is poor, but Bailey IS getting more SOG per game this year than in any prior year as an Islander - over 2 shots per game when his career #s used to be 1.35 - that's a huge increase which should lead to more goals. It hasn't done so mainly because Bailey's shooting % is half his career rate, a trend which won't continue.

In short, Bailey's shooting more like Isles fans have always wanted! This is going to be good!

In addition, Bailey's actually adept at offensive zone entries (Neutral zone play) with 65% of his entries coming with possession (mainly by carry).

However, Bailey's #s do have some issues. While the basic possession #s (Corsi/On) aren't bad, they're coming with slightly favorable zone starts. Moreover, the team is clearly performing better possession wise with Bailey off the ice, and while part of this is because he's never been with JT, it does suggest the possession #s are solid not because of Bailey but in SPITE of Bailey. Remember that Lubo has been the best Isles' possession player since joining the team, and unlike the other Isles, Bailey has played all but 2 games with Lubo in the lineup.

So Bailey remains an enigma - possession wise he's not what we'd hope for for a second-liner, but the goal scoring should increase IF he can keep up this rate of shots. That'd be nice.

Nielsen's #s are arguably the most interesting of the three 2nd liners because Frans has escaped the criticism leveled at Okposo and Bailey for the most part. But well, these numbers look bad. How can we explain them?

First, let's clarify what I mean by the numbers look bad: While NIelsen is facing tough competition (very tough in fact), it's nowhere close to a competition level that should make his results look horrible.* In fact, Nielsen is facing easier minutes than previous years because he's no longer starting most of his shifts in the D Zone - he's actually facing more offensive zone starts than D Zone starts. Yet the team is being outshot by around 10 shots per 60 while Frans has been on the ice. That's bad.

*Statistical Tangent: The impact of Quality of Competition on stats is apparently vastly overrated. While quality of competition faced clearly matters - if a top line faced only a line of goons they'd have like a 66% possession rate - over the course of a season the gap between players' quality of competition is basically minimal. Two reasons for this should be obvious: 1. Changes on the fly make it impossible for players to match up perfectly with opposing lines resulting in each line getting a lot of time vs every other caliber of opponents; and 2. the only real time teams can match lines based upon competition is with last change, and each team will have 50% of their games without last change, negating that impact. For more, read http://nhlnumbers.com/2012/7/23/the-importance-of-quality-of-competition.

Of course, it's not as bad as it looks. See Nielsen started the season on a line with Boyes and Grabner. In their 8 games together, they combined for a corsi of -40 - which is pretty ugly. They played more defensive minutes together than the #s above would show, but still, ew. After this Boyes was switched off Nielsen's line, and FN-GO did a little better, but was still poor. Four games later, Nielsen was separated from Grabner.

The end results was that Nielsen suddenly became able to win the possession battle - his zone-adjusted corsi % is around 51% without Grabner - meaning the Isles took 51% of the shots while Nielsen has been on the ice without Grabner. That's in line with Nielsen of the past. Of course, Nielsen's #s were so bad with Grabner and Boyes that the stink of those #s still makes Nielsen's overall #s look bad - though they were a lot worse through 13 games!

What should be done with Nielsen going forward?

Well, despite his poor performance with defensive minutes earlier this year, I think it's clear Nielsen should be put back in such a role, even if his linemates are kept the same. We have 3 years of data showing Nielsen to be a plus possession player in defensive minutes; those years carry more weight than a 1/4 season of data. Nielsen should be put back into the the defensive but not too defensive role he was carrying previously.

Moreover, It should be noted that Nielsen, like his linemates, has a shooting % roughly half of his career rate. What this means is that his ENTIRE LINE is due to regress in the good way - and they'll start to put up more goals as the season goes on. This should help the secondary scoring argument, and would offset a loss in offense that they might get by shifting Nielsen-Okposo-Bailey to a more defensive role (though I'd think this would help the O anyhow by shifting more play out of the D Zone).

Grabner is another controversial figure among Islander fans this year, but for different reasons than the above 2nd liners. His shooting and goal totals have resulted in fans wanting him to play more, not less, and his PK (#s not shown) has been excellent.

That said, he was a part of that same awful Frans line to start the year and the possession #s were not pretty. For some reason, the two didn't work together, even without Boyes for 4 games. Was this a fluke? Well, I'd say the overall awfulness was fluky, but lost in last year's #s was the fact that Nielsen was better without Grabner while Grabner was unaffected really by having Nielsen on his line. They were nowhere near as bad as they were this year together, but it suggests that FN-GO was really a fluke in the pan two years ago.

Now Grabner was split off from Nielsen on game 13, but unlike Nielsen, who gained a more offensive role, Grabner was deployed in the defensive end, mainly with Marty Reasoner. And despite what you might think...this has REALLY REALLY WORKED. Despite Defensive minutes, Grabner has a corsi % of over 52% with Marty Reasoner - they're starting with the puck more often than not in the defensive zone and yet still decently outshooting the opposition. That's fantastic!

Moreover, it's not like Reasoner has hurt Grabner's shooting - Grabner has gotten more shots off per minute with Reasoner than he did with Nielsen (and more goals too). Simply put the Reasoner-Grabner pairing has been very successful and probably should continue, even if Grabner-Nielsen is still the norm on the PK..

This doesn't mean by the way that Grabner should get this little ice time by the way. Cappy should deploy Grabner more! Two ways he can do this:

1. Double shift Grabner onto the 3rd line, especially if Boulton is in the lineup, particularly late in games.
2. Play Grabner-Reasoner more! I know fans dislike Reasoner (More on that later) but this line has been really effective...use it more!

A FINAL NOTE OF CAUTION. As Grabner did 2 years ago, Grabner's shooting % is fluky high. In short, he's due for regression - we should expect a 13% shooting Grabner going forward rather than a 16% shooting Grabner. This'll slow his goal scoring a little. But extra minutes (and thus more shots) would help limit this.

Marty Reasoner had a really good defensive year in Florida in 2010-2011, which he parlayed into a two year deal with the Isles. And then last year he was miserable. Just miserable in every sense of the word.

But this year, without most fans noticing it, he's really really rebounded. And if you look at last year's #s, Reasoner's #s aren't nearly as bad, though they're still bad, when he was without Jay Pandolfo. It seems Jay Pandolfo may have been a plague on that line.

By contrast, armed with Michael Grabner and a random linemate on his wing, Reasoner has done really well this year. He's facing tough defensive minutes - nearly 3 defensive zone starts for every 2 offensive zone starts - the Isles are basically even possession wise (just below) with Reasoner on the ice, and as covered above, they're well above even when he's together with Grabner.

Hell, even the shooting shouldn't be looked at THAT poorly - Reasoner may no longer be a 9% shooter, but he's not a 0% shooter either. My guess is he ends this season at around 5% shooting, with him putting up a few goals here and there (say 3 in the remaining quarter of the season). But for these minutes, that's all you need, and as noted above, he's not seeming to hinder Grabner in any way.

I've got a theory on this actually - Reasoner's clear specialty, even last year, was the ability to win Defensive zone faceoffs. In fact, while Reasoner is on the ice, the Isles have won 43 out of 70 defensive zone draws (61%). That's fantastic. Once the draw is won, the puck can find its way to Grabner, who can race the puck out of the zone and create a quick scoring chance. Boom, defensive zone cleared, offensive zone entered. I'm sure this exact scenario is actually quite rare to happen as i just detailed it so simply, but you get the idea - win the faceoff and Grabner presents the threat to get the puck out of there as soon as the puck gets to his stick.

Really Reasoner's only fault is not getting the puck in to the offensive zone by carry enough. But it's not as bad as it could be really.

In short, I'm serious - Play Reasoner-Grabner-X more please. That's a good line. Give it more of the D zone draws that you would give to JT and give JT more of the o zone draws in their place.

David Ullstrom was clearly a guy Capuano did not like - frequently benched in favor of Boulton, given the least ice time when he did play and now sent down to the AHL. But the #s here aren't really that bad. It's not clear if he was a good fit next to Reasoner and Grabs, but his possession numbers arent too negative despite tough minutes, he was good at carrying the puck into the offensive zone, and his shooting #s were certainly going to increase given his career shooting % in the AHL was around 13%.

That said, he was committing too many penalties, and that along with the occasional gaffes were too much for Capuano to handle. I don't think that's the right idea - he's still 23 and can learn best in the NHL - but it's not like he's dominating either. There's no way in hell sitting him in favor of Boulton was the right move, but we'll get to that. Hopefully Ullstrom will be seen again this year.

Casey Cizikas is another young kid getting his first real exposure to the NHL this year with the Isles, but unlike Ullstrom, he's found a way into the lineup every game after a few early scratches. Cizikas has been centering what traditionalists would call the "checking line" - personally I'd call it either the secondary defensive line or the defensive 2-way line. What I mean by that is that Cizikas is getting more D Starts than O Starts this year, a sign that he's playing a defensive role, but he's not being buried in the D Zone or really facing too few offensive zone starts - that role belongs this year to Marty Reasoner.

Possession wise, this is a bit of a meh profile. Cizikas - particularly with Martin and CMac - is losing the possession battle and the tougher minutes don't excuse this fact. His rate of zone entries with possessions has dropped, which isn't good because the other two guys on his line can't carry it in either. Right now, if we separate Frans' awful start, CC looks like the team's worst center at driving play, which means less minutes for JT in the offensive zone. Not what you want. Mind you, again, CC is a kid, and should improve. But he's not really deserving of the accolades I see occasionally bestowed upon him.

Offensively, Cizikas has done sort of well so far this year. His shooting % is roughly the same as you'd expect from his AHL/NHL #s, so it's not clear it'll regress despite the fact that 14% is quite high. However, he's getting less than a shot on goal per game, which is really disappointing - Reasoner and Grabner are getting less minutes and facing harder defensive minutes and yet Reasoner is getting more SOG. CIzikas needs to get more pucks on net - after all he averaged over 2 SOG per game in the AHL.

Martin's rookie season was really bad. I wrote a post talking about how he was extremely bad, to much outrage, and it noted how the team was outshot by around 20 more shots per 60 with Martin on the ice than when he was off the ice. That's horrible. Martin's sophomore season was much better - the Isles were only being outshot by 5 more shots with him on the ice than they were with him off the ice - acceptable for a 4th liner playing defensive minutes.

This year, despite a start looking close to last year, Martin has regressed a little bit. He's not rookie year Martin, but his possession #s show he's not doing a good job this year. The Isles are being outshot by 9.7 shots per 60 with him on the ice, and are actually over 10 shots better with him off the ice. Martin's still playing the same defensive minutes so that's not an excuse here. Moreover, his offensive output is not particularly impressive - he's got the same low shooting percentage as he has had in his career and he dumps or tips the puck into the zone nealry 4/5 of the time he causes a zone entry. That's not a way to create effective offense.

Moreover, unlike Frans Nielsen - whose #s are clearly the result of a flukish bad line earlier in the year - there's no evidence that this downturn is caused by simple bad play with a few teammates. The only guy whose #s Martin has improved when on the ice together with Martin (minimum 20 minutes with Martin at EV) has been Eric Boulton (and those #s are still awful) - everyone else has gotten worse when playing with Martin than without him.

I don't know why this is - perhaps Martin is again being too overzealous with hits and being out of position; maybe more likely it's because the unrelenting amount of dumps crippling the Isles ability to hold the offensive zone with Martin out there on the ice. After all, in the 10 games I had with Martin last year. his entry rate with possession was 49%, nearly double what it is this year.

I admit, I had very low expectations for CMac this year - a journeyman AHLer who's nearly 30 already does not inspire confidence when brought up to play major NHL minutes for the first time in his career.

For what it's worth, the #s are okay. CMac is facing tougher D Zone starts somehow than anyone on the team (he's spent some amount of time with Reasoner, which is the cause of this), and the team is being outshot by 7 shots per 60 with him on the ice. The team's 8 shots better with him off the ice in fact. Those aren't good #s by any means, but for a "4th liner" picked off the scrap heap, they could be a lot worse. The frequent dumping is also a problem, but not to the extent of Matt Martin.

In fact, if not for age, I'd think the #s would say the Isles should keep CMac and get rid of Martin if they wanted a tough grinder going forward. Age is of course a major factor so I'd not suggest this (Martin is likely to get better, CMac is going to get worse). But even offensively, CMac is superior - that 11.4% shooting % is probably fluky (Career in the AHL is 9.48% with one extremely fluky 16% year) but he's getting a reasonable amount of shots on net (more than Cizikas somehow).

In short, I won't weep for CMac when he's gone next year, as he likely will be. But I'm not too upset he's playing either.

Aucoin is even more of a journeyman than CMac, and almost certainly won't be with this team next year due to age and not great performance. But honestly, he's been underrated. His possession #s are nearly even - the team has barely been outshot with him on the ice - and he's played tougher defensive minutes than most of the other forwards. The shooting was enormously fluky at the start, but his possession #s are solid and well, they've been solid throughout his career. If we were a contender this year or were a stanley cup contender next year, I'd recommend keeping Aucoin.

However he's 34 (going to get worse) and is going to take up spots needed by other forwards if he's on the roster next year and will have to play out of position on the wing. If we were to waive Aucoin, he could be a helpful 4th line center for a contender. For this team however, he's not really useful.

The Isles are being outshot by 27 shots per 60 with Boulton on the ice. TWENTY SEVEN PER SIXTY MINUTES.

Goons are bad, mkay. Friends don't let friends play goons.

--------------------------------------------------------------------------------------------------------------------------------

Conclusion:

I think I've said my own conclusions in this piece above, but I'll probably wrap up this series with a short 4th post sometime in the next week. I hope this was informative.