AnalysisStats To Chew On

A Mixed Bag of Mediocrity

With one game left in November, the Rangers have played 25 games and sit 3rd in the Metropolitan division. While that of course exceeds expectations for many, if you were to choose a statistical category that the Rangers rank almost that high in, what would it be?

Save Mr. Lundqvist in net, the rest of the team ranks range from average to downright bad. Not that this wasn’t expected, and injuries have certainly played a role in dimming any hope of over-achieving, however this November run has been buoyed not by an improvement in play, but rather by some fortunate shooting and consistent play in net by Lundqvist.

That all being said, as we inch towards game 30 we are reaching the point of the season where underlying metrics such as Shot Attempts (Corsi) and Expected Goals begin to have their highest predictive powers for the season. I wrote in early September that the majority of models had the Rangers just under 80 points, and after banking more points than expected in November, the Rangers now range anywhere from 82 to 86 points from the public models available. Last year, that would place the Rangers anywhere from 19th to 22nd in the league standings. Is that what the Rangers are this year? Barring any major personnel changes, the answer to that question is likely yes, especially given David Quinn’s early choices on how to deploy his lineup even before the injury bug hit certain players.

The reason the answer is a likely yes to the Rangers being just below average (rather than full-on tank bad) is due to two reasons:

  1. They hang with teams when it comes to quality shots and chances, for and against
  2. They consistently get out-attempted in shots (Corsi), indicating that they’re still in their own zone much too often.

The first item is a good problem to have, to an extent. While roster construction (aka talent) certainly drives a team’s expected goals for the most part, there is also a link to coaches who prefer different styles of play. In recent history, the general issue with NYR hasn’t been their Expected Goals for per 60 (xGF/60), but rather their xGA/60 (against), which ranked dead last in 2017-18. Under Quinn, high-danger and medium-danger chances are down (slightly), but because they’re still struggling with the shot share, low-danger chances are up (via NaturalStatTrick, see below). Speaking relative to last year though, this is of course an improvement, and hopefully protecting the middle will be a staple of Quinn’s system.

In the end, though, the Rangers are spending too much time in their own zone, which can be seen in their 29th-ranked Corsi-For percentage (only Anaheim and Ottawa are worse). From 2013-2018 in the NHL, the correlation between Corsi For and Expected Goals is about 78 percent. Only 15 out of 79 teams who have posted Corsi-For percentages under 50 have ended up with an Expected Goals For % over 50 (aka 19%). Unless NYR also begins out-shooting teams, their xGF % will likely fall as well.

The shot share issue is also why you may have seen some of NYR Twitter slightly upset as to how Quinn is deciding who plays, and how much. While the Rangers are certainly on track to eventually sell on some players in order to continue the rebuild, it should be stated that some players are not being scratched or penalized for their play like others. Let’s take a look and some Forwards first.

The way to read this graph is to know that up and to the right is good, while down and to the left is poor. With this, we can see that Hayes, Kreider, and Chytil have carved out a niche for themselves, regardless of the Rangers opponent. I think most would agree, even without knowing what Expected Goals or Corsi is, that those three have played well. The middle is where it gets a little muddled. Zibanejad is always tasked with playing the other team’s best players and Zuccarello was riding shotgun with him before his injury. Namestnikov and Fast have had up and down starts, and Buchnevich was just finally getting going before his broken thumb.

The real focus, if NYR were to actually try to get better with some addition by subtraction, would be in the bottom left hand corner. Spooner was traded away (months too late) and while McLeod is injured, many would agree he shouldn’t be playing regardless. Our junior saviors Andersson and Howden are interesting cases. Andersson hasn’t had much of a chance up with increased time on ice, so that is something I wouldn’t mind seeing. Howden, however, is masking some not-so-great play with a high shooting percentage (16% personal, 10.7 on-ice aka him and his teammates). A brief stint in the AHL may not hurt him in the long run.

This finally brings us to Mr. Jimmy Vesey. On pace for a 26 goal season at age 25, his situation screams “sell high” to me as we move forward. Vesey has never been a positive shot share (Corsi) or Expected Goal player relative to his teammates in his 184 NHL games played. And while he clearly has a nose for the net and some shooting talent (career 13%), it is not necessarily at a level that couldn’t be replaced with the cap space that will be available to the Rangers.

As has been evidence the last few years, the Defense is where Jeff Gorton has his most work to do. To start positive, Fredrik Claesson has rightfully received praise from many outlets, and is doing his work while mostly playing against middle-six forwards (based on opposition TOI over the season). His main D-partner, Kevin Shattenkirk, has also had a nice start to the season despite maybe not looking as visually pleasing at some points. One item that I haven’t seen highlighted much with him is his career-low 2.1 shooting percentage (career average = 6.3). He’s still getting about 2 shots on goal per game despite not being on the Power Play as much (which, why?), so unless he’s going to have a snake-bitten season I think 22 may begin to find the net eventually.

Marc Staal, Adam McQuaid, and Brendan Smith all fall into the “fine” category as much of the same goes for them as Shattenkirk: it just hasn’t looked pretty at times. To me, this is likely because of the point of their careers – all three could probably be ok to good 3rd pair players on a very good team, however the Rangers are not that. Brady Skjei is having a bit of an extension-signing hangover start to the season, so it will be curious to see if he rebounds as he’s played less minutes in the last 10 games or so. Anthony DeAngelo has played well relative to his teammates, but he’s still below water overall. That being said, it will be nice to see him stick in the lineup as he for sure creates chances and is on a contract year as well.

As with the Forwards, there is a single name at the end to focus on, and that is Neal Pionk. Pionk is and/or does all of the following:

  • A quick moving skater
  • Makes impressive stretch passes
  • Has a shot percentage like a forward (9.8)
  • Only has a 41% shot share and is noticeably lower than his teammates
  • Is still inexplicably strapped to Marc Staal, one of the Rangers worst defenders in recent memory

With only 4 of his 15 points coming at even strength (none shorthanded, mind you), he just isn’t playing at a level yet that indicates he is or can be an effective NHL player. This could have something to do with Staal, sure, but he shows similar numbers away from Staal as well. Time will tell, as it seems Pionk will be a staple of the NYR lineup this season.

In the end, the Rangers have been a mixed bag thus far this year. In the beginning, when the wins weren’t coming as easy, there were some indications that a healthy roster may be better than most projected. Then, as the injuries hit and the wins started falling with hotter shooting and consistent goaltending, the play also started going south despite the points in the standings. We are likely still headed towards a decent sell-off at some point from Gorton, but it will be interesting to see if anything changes as the look of the lineup develops as well.

Show More
  • The stats are what they are, and based on what I was expecting this season, I’m fine with it. I just love the hustle that we are seeing on the ice, the emotion that was lacking under the other coach, and the development of our kids. I give this team another two years before we will be considered a true cup contender!!!!!!

  • Thanks Rob, everything you show here is supported by the eye test. The Rangers very rarely have the puck for any length of time.

    Lack of skill.
    Lack of skill.
    Lack of skill.

    But let’s re-sign Hayes for the next 6 years. He’s not the problem because he possesses the puck too much, but he also can’t be your best player either.

    • Appreciate it. I’m leaning towards the trade Hayes opinion. Mika Z. is a full-on #1C with a solid contract at age 25. Chytil can clearly play and I don’t think it’s going to matter if you move him back to the middle, so there’s your 1-2 punch. Howden can also likely play, but just not up in the lineup for probably a couple years. Lias is a work in progress.

      Teams do well with depth at C, but if Hayes’ shooting % comes up and he pots another 20+ goals this year, he’s going to get paid and I think that money can go to use elsewhere among the lineup.

        • You should probably write it, Mint. Your semiliterate, loudest-drunk-at-the-bar “writing” style is best suited for fiction as it is. And the idea of Kevin Hayes being a Ranger after this season is fictitious. So pack your article full of your customary gibberish, leave established statistical evidence out, and repeatedly parse comments made in one article by Gorton six months ago like you like to do. Before you know it you’ll have 750 words on this topic.

      • The Rangers cannot tie up $6M per on a 50ish point player at this point. They need to clear cap space to bring in guys like Panarin and Stone who have superior offensive skills, not to mention a true top pair D man.

        Bringing in guys like that will make the younger player better. And Zib needs to stay as he is somewhat of a #1 center and he is also RIGHTY, a trait the Rangers have very little of throughout their line up.

        • As if Panarin and/or Stone will command anything like 6/7m per … try about 8 — and I think Panarin might be closer to 8.5m. I could be wrong but I doubt it, they’re the two premium FAs this summer. Sure I would want 1 of those 2, but that doesn’t mean I would automatically want to get rid of Hayes if he’s asking $6m.

          The grass is always greener …. move Namestnikov at $4m and keep Hayes at circa $6m, then signing either Panarin or Stone is still feasible. Howden, Chytil and Andersson haven’t proven they can be reliable game in and game out Top 6 centers, let whomever of the 3 grow into it — maybe Chytil is just fine on the wing for now, he seems to have taken to it (even his defensive game). Just my opinion. 😉

          • Oh, Panarin and Stone will combine for $17M, I know that, and they will be worth it.

            As I have said over and over, this is the best we will get from Hayes, it goes downhill from here. And he’s never reached 50 pts so far.

            The Rangers have him now, how they doing?

          • You do realize he’s the virtually the same age as Stone and a year and a half younger than Panarin. What’s makes you think they have better days ahead compared to Hayes? Besides,. Stone has never cracked 65 points in a season, is he that much of an upgrade over Hayes? Anyway, those 2 are wingers … Either one needs to slot in for a winger — so subtract Zucc and Namestnikov and insert Stone or Panarin. Hayes isn’t the right player to sacrifice in this instance.

    • “but he also can’t be your best player either.”

      ^ It’s funny because I have the same feelings about Zib.

      I wouldn’t be to hard on Brooks, he’s just reporting on what he hears.

      • I think that they are “using” Brooks to beef up Hayes’ trade value.

        Look, we could be sitting here in Feb with an extended Kevin Hayes, but if that happens it will be another colossal mistake IMO. The Rangers need to upgrade at the top of their lineup, on both the offensive and defensive side of the roster.

  • I hope we keep Kreider. He’s the one I want to see remain a Ranger and even with a young rebuilding team, you need veteran leadership.
    If they decide to trade Hayes, Gorton has better hit one out the park. The kid is young still and is playing like a beast. Teams interested must be drooling watching him skate effortlessly through defenses with the puck. He’s a big center that skates well. A perfect 2C on a contender. I honestly wouldn’t be opposed to retaining him long term. He may want to go to Boston tho. Time to look a Bruins assets.

  • Thanks Rob. Excellent as always. Before this season when I looked at this roster I saw a team that was going to end up in the 9 to 15 range, rather than bottom 3 with a chance for Hughes. I’ve posted elsewhere that one of my main conclusions seeing this team so far is that the FO really did not understand why the team bottomed out last year. In my opinion it was due to three reasons:

    1. Poor roster construction. The team had no depth and was not well designed, especially the forwards;
    2. Injuries. Once injuries hit that thin lineup, our team absolutely collapsed. We weren’t good heading into the New Year, but you saw a complete collapse when Mika got injured; and
    3. AV being AV.

    What’s funny is that the FO inadvertantly solved 1 and 3 which allows them to overcome problem # 2. The team has a good and deep forward group, and while the defense is not talented, it has depth (they even got lucky they picked up Claesson). So we can actually overcome injuries (one point I would also make is that our recent spurt coincides with missing two top six forwards in MZA and Buch!).

    On number 3, in your post you note how the QA has dropped dramatically. All one has to do is look at the heat maps of where shots come from to see that the team does not really give up shots in the slot and royal road shots. Why? Because of some tactical shifts from Quinn and being focused on it (Dave Shapiro wrote about some of the dzone tactics).

    This plus the fact that the FO never really understood how important Hank was to this team. With the drop in quality against, Hank can mop up lower chance shots pretty easily as long as he gets a clear view. But as Rob noted in his piece, the rates against are still a disaster, and it means that the possession of the game is controlled by opposing teams.

    Having said all that I think we should be worried about some of Quinn’s deployment choices. Leaning on a Staal-Pionk pairing really makes no sense to me. He’s getting a ton of out of a lot of forwards, but McLeod being on the ice is a joke to me.

    And of course we should be worried about Gorton who is in the lower third of GMs. I think we have a good core and we can compete sooner than we anticipated. But is Gorton going to do something crazy? I really hope he learned his lessons from his botched Stepan and McD trades (the fact that this summer we were hearing rumors about the Rangers looking at other players in that draft other than Lias means that the FO which never leaks was worried about the perception of that horrible trade). You don’t sell good players in their prime for spare parts. That should be the mantra here. I also hope we don’t bridge Buch as that would be dumb.

  • I mentioned in another comment that if you wanted to know whether a team had scored more goals than given up, counting shots was about as good a predictor as flipping a coin. 14 of the 31 teams either outshot the opposition and scored less or were outshot and scored more. PDO is far more important than Corsi, no ifs, ands, or buts. Statistically, the Corsi numbers are far more accurate. Sample size error makes measuring PDO something of a crap shoot. However, things can both be important and difficult to assess.

    Addition by subtraction is not called for here. The Rangers do not have better players than the ones that they are using. McLeod should not be playing in the NHL ideally, but he is better than Lettieri and probably everyone else in Hartford. It wasn’t even a plus to get rid of Spooner – it was a plus to get Strome.

    Marc Staal is a good defenseman, likely still better than Brady Skjei will ever be. At this point in his career, he is a quality second pair left defenseman, who is actually holding his head above water playing first pair with a guy whose defense is suspect. Staal is not holding Pionk back. As you say, the book is still out on Pionk. But his steady pairing with Staal is giving him every opportunity to develop into the best dman he can be – and that is what a rebuild is about.

    The current team is a mixture of good players, prospects, and filler – about what you would expect. Quinn seems to be doing a good job of transitioning the true prospects into the game and not letting the filler get in the way. But one word of caution. Good players are far to come by. The fact that a #7 pick may not amount to much indicates that prospects don’t guarantee anything. You don’t just trade good young players like Hayes without a clear plan for replacing them.

    • PDO of course influences goals for and against results, but it is a correlation, not a causation. This is because PDO is not reliably repeatable year over year (less than 15%). Expected Goals and Shot Attempts/Corsi For predict future goals for at nearly double the rate at this time of the year (30-35%, and newer models are slightly more accurate):

      Teams will see their xG and Corsi change the rest of the way, of course, but it will not change as drastically as PDO can, because shot and save percentage swing more dramatically from a statistical standpoint.

      • No, it is a causation. And what is this time of year crap?

        Life is very simple really. Number of goals equals shot attempts x percentage of shots made. Ditto for goals allowed. If you look at real data over an entire season, you see that the second factor is much greater than the first. And it is simple common sense that whether or not a shot goes in is simply not luck. Some players have better shooting percentages over their careers. Some goalies are better at stopping shots. Some defensemen are better at giving their goalies easy to stop shots. Some systems (like AV’s) produce better numbers. It does not all average out.

        It is damn hard to reliably measure “true PDO” based on 20 games – that is undeniable. Just like the fact that Derek Jeter starts a season 6 for 20 gives little indication how he will end up. EXCEPT you also know that he is Derek Jeter and you have career data. And if Marc Staal has been a good PDO player for years, guess what?

        Now I have not yet researched xGF%, which is of course an attempt to integrate PDO into the shot data. But it is mote subjective, since “good” shots are not purely objective.

        But the key to that 30% number is that is all that Corsi is attempting to achieve. It ignores the other 70% of the game. And leads to bad judgment.

        It may be obvious that 30% is better than 15%, but it is not actually true. If you are going to make decisions based on 30% of the information and I know precisely what information you have, I can outsmart you.

      • If you read the article you cited, you can see how to do better. Stamkos is considered and there is a model for regressing his shooting percentage to the mean. But there is no reason to expect this. We have career numbers for Stamkos that are much more accurate than one season numbers and we expect Stamkos to regress’s to those numbers and not simply numbers for all forwards.

        If you look at a single year and notice Stamkos has a higher shooting percentage than Tanner Glass, one might expect both to regress to average forward numbers. But we absolutely know that neither player is anything close to average and regression is toward some other number.

        In some years, Stamkos will be both above average and below his career norms. In those years, you expect an improvement, not a decline.

  • We have excellent second tier talent. This is why we are hanging around. Absent of a stud top line and a stud defenseman, we are a good team. The problem is that good teams do not go far in this league without a few special players. I am not putting any of our kids as budding superstars yet, so that talent needs to be bought. We probably need $8M a year each for a top defender and a top forward (give or take a few million). If the only way to acquire that is to shave $7M off of our current roster, then we have to move players. Could be Zucc, or Hayes, or Names, but we need to clear Cap space. Gorton has to know what he needs to turn this club into a top-tier team and it isn’t Kravsov.

    • No sure-fire superstars yet, but possible highly skilled to elite players in 19 year-old Chytil and 18 year- olds Kratsov and K’Andre Miller. If they can grab a stud forward and a stud defenseman it will likely have to be a free agent because the team is not going to tank. Yes they may have to clear some cap, but they are in pretty good shape at the moment to make moves if opportunity knocks.

      • Trade for picks that give you CAP space or for a disgruntled stud who needs a change of scenery. All the rest of the cash for Panarin as a free agent

  • I still believe that Gorton signs Hayes if he doesn’t have to give him a NMC/NTC. BUT, I think Hayes will insist on one and Gorton will trade him. Giving Hayes that type of contract is just bad revisionist history, so JG will try to avoid another Stepan situation.

  • Back to top button