November 24, 2015

Are the Rangers lucky or good? Why the best team in the NHL is being questioned

2015 PDO to WINS
Nash lights the lamp again (AP Photo/Julie Jacobson)

Nash lights the lamp again (AP Photo/Julie Jacobson)

The Rangers have 16 wins and 34 points in the standings after last night’s win against the Nashville Predators. They along with the Dallas Stars and the Montreal Canadiens are tied atop of the NHL mountain. But why are people still questioning the legitimacy of the Rangers?

Digging a little deeper, the Stars have 1 more win (17) in the same 21 games played and the Canadiens share the same goal differential of 28 but with 1 more game played. The Stars and Habs goal differential is due largely in part to their goals for, while the Rangers goals against is the best in the League by far.

Yet, the Rangers are being labeled “lucky” even though they have gone to 3 ECF’s and 1 SCF in the last 4 years. Winning the Presidents’ Trophy last year obviously didn’t sway opinions, so why should being one of the best now change that? Well, because apparently FANCY STATS can’t explain them.

RANGERSTOWN

RANGERSTOWN

Listen, in Rangerstown CORSI is dead. You know why? The Rangers killed it! For years the Blueshirts haven’t been one of the top teams in “shot events” that claim to measure possession. Last year, the Rangers ranked 19th in Corsi (49.5%) but finished #1 overall in points. They lost in the playoffs because Mats Zuccarello almost died and 4 defenseman were playing hurt with another (Kevin Klein) coming off a broken arm. Not because of corsi events catching up to them.

The year prior, they made it to the Stanley Cup Final with a solid Corsi of 52.4%. That was good for 7th in the NHL CF% wise but just 12th overall in the points standings. In 2011-12 they made it to the ECF but finished that year ranked 23rd in Corsi at 47.7% (2nd overall points in the NHL). And just to complete the 4 year span, the 2012-13 Rangers that got bounced by Boston in round 2 had a corsi of 52%, good for 9th in the League (12th overall points-wise).

So even though the Rangers have broken Corsi, they are still being doubted. Even though they sit 2nd to dead last in CF% at 45.5%, that’s not what people are pointing to anymore. Now it’s all about PDO.

What is PDO? That’s your On Ice Save Percentage (which should be renamed your Team’s Lundqvist Number) plus your On Ice Shooting Percentage at even strength (5v5). That means you take the Rangers current SV% of 96.25 and SH% of 11.14 and you get a PDO of 107.39! That is almost 5 points higher than the Senators who are in 2nd. That’s 6.2 and 6.3 higher than Montreal and Dallas respectively. Here’s the problem, analytics people view very high PDO’s as lucky and that it’s an unsustainable form of winning. Here’s what I mean via a blogger on Twitter.

Notice the upper right hand corner? See what it says there? “Too Good To Be True” …well that’s what most people staring to long at spreadsheets and charts would like you to think too.

Take last night’s game for instance. The Rangers were outshot badly for two periods and could barely generate offense. Why? When you have to kill 8 minutes of penalties in the first 21 minutes of a hockey game, it’s hard to get going. The Predators kept coming off the momentum but the Rangers did a good job of keeping them at bay and Henrik Lundqvist was being what he is…elite. And no, I won’t apologize for having the greatest goalie in the NHL, just like Montreal doesn’t have to apologize for Carey Price.

henrik-lundqvistThe Rangers stuck to their system last night, and in the second period generated an odd man rush that led to a Rick Nash goal. His 4th goal in two games, indicating his measly 2 goals in 20 was just as important an unsustainable number. Then the Rangers got their first PP of the contest in the third, generated offense and eventually scored and dominated the remainder of the game for a 3-0 win. Their SV% was a perfect 100% and their SH% was 6.33%. Which means a PDO of 106.33 for the game. I see a pattern.

Now for the shocker! I bet most of you reading this were getting ready for me to annihilate PDO. Well, I’m not going to but I am going to do is put it into some context with the Rangers.

Just like I did for when I wanted to see if Corsi and Fenwick correlated to winning, which it doesn’t, I tested the same for PDO. Mind you, I was expecting better numbers because we are talking about stopping the puck and putting it in. To be honest, I was surprised at just how strong the correlation actually was.

Below are two models for the last 2 seasons and how higher PDO’s correlate to wins. The numbers were impressive. First let’s look at last year and you will see many blue dots close to the line explaining nearly 50% of variation in the numbers (r2=0.4719).

2015 PDO to WINS

2015 PDO to WINS

Then just to check one more full season, I did the 2013-14 season and the correlation was even stronger at almost 62%.

2014 PDO to WINS

2014 PDO to WINS

The takeaway for me is simple, PDO is a great way to assess how good a team is playing. My issue is why are teams that are greatly outperforming others considered lucky? In this case the Rangers.

As I pointed out before, the Rangers CF% numbers were all over the map since 2011-12 and yet they were a top team. When you look at their PDO it tells a different story, one that aligns better to their actual place in the NHL standings.

2014-15 PDO: 101.9 (1st)

2013-14 PDO: 99.7 (19th)

2012-13 PDO: 100.9 (3rd)

2011-12 PDO: 101.0 (6th)

As you can see the Rangers are consistently good at PDO. The low 2013/14 season can easily be explained, because for the first few months they were learning to play under a new coach and system.

Now I can tell you this also, the Rangers will not finish anywhere near 107 in PDO this year. Matter of fact, I expect a regression just like everyone else between 100 and 103. That doesn’t mean the wheels are going to fall off the train the Rangers are rolling on now either.

This team, for all its issues that some people feel the need to harp on is still better than almost any other team in the NHL. Do the Rangers need to tighten up defensively and cut down the other teams shots? YES! Not to improve corsi numbers but to increase their odds of winning, especially in the playoffs when games are tighter and scoring is at a premium. The Rangers also need more offense from their 2nd and 3rd lines. Right now, they are being carried by their top line and that will be checked to hell come April in the post-season.

Bottom line is this, the Rangers being a top PDO team is par for the course…not because they are lucky but because they are good.

Henrik Lundqvist is an elite goaltender probably playing the best hockey of his career. However, Cam Talbot and now Antti Raanta have found success in NY as well. Luck? Or do the Rangers have a system and the right personnel to fit it in place? To me the answer is they are a skilled group that understands what the coach wants and adheres to the system. Furthermore, they know how to win and not crack under pressure. This was obvious last night for 40 minutes against Nashville. These are things no spreadsheet or chart can quantify but they exist.

John Minchillo/Canadian Press

John Minchillo/Canadian Press

Recently, Alain Vigneault gave his thoughts on advanced stats on the MSG Network show, The AV Squad 15-16. Here is what he said, “There’s a lot of theories right now, out there about puck possession, time in the offensive zone, Corsi, Fenwick, etc, etc. I think you got to take that with a little grain of salt.” Vigneault then continued to explain,”What you’re seeing on the ice as far as coaches and what we analyze as far as scoring chances etc…I think gives us a real good indication of how the team is playing and how players are playing.”

Then AV committed the cardinal sin of the analytics world when he admitted to relying on the dreaded “eye test”. His exact words were, “You can’t over-analyze with those numbers. Sometimes you get a wrong feel and I prefer trust in my experience and what my eyes are seeing.”

So now the coach of one of the best NHL teams over the last few years has gone on record as I have that these numbers aren’t always trustworthy. AV, like myself and many others still like to rely on what we are seeing primarily with our eyes to make the final assessment on play.

Now when it comes to PDO and how the Rangers are “lucky”, I’ll just use the coach’s words and say, “sometimes you get a wrong feel.” As far as those PDO charts and how they are interpreting this Rangers team as lucky…the coach is spot on that you can’t over-analyze it.

And when it comes to the New York Rangers…they aren’t lucky. They are just that damn good!

 

Share