Nashville Predators Twitter defends Roman Josi

Nashville Predators defenseman Roman Josi (59). Mandatory Credit: Aaron Doster-USA TODAY Sports
Nashville Predators defenseman Roman Josi (59). Mandatory Credit: Aaron Doster-USA TODAY Sports /
facebooktwitterreddit

Now that Shea Weber is gone, where does Roman Josi fit on the Nashville Predators depth chart?

Wednesday saw what seemed like the entire Nashville Predators blogosphere (including myself) arguing with a man on Twitter over Roman Josi. I personally have never seen such an overwhelming outpour of rage by Predators twitter defending a player. Even the “Shea Weber is average” drama the last few days only brought up a dull roar from the electronically savvy Nashville Predators fans… more on that later.

But it was a sign that the two popular methods of evaluating players; analytics, or the old “eye test” are coming to a boil in Predsland. Who is right? As with everything in life, the correct way to do things probably falls somewhere in the middle.

Here is the tweet that stirred the pot.

This brought out an almost immediate response from the Nashville Predators blogosphere (that’s a word, right?). Somehow this tweet was seen and everybody came out of the woodwork as if Chris Thomas was trying to burn Bridgestone arena to the ground. I am sure part of this was intentional trolling, knowing full well what kind of response he would get. I can’t throw stones, I took the bait as well.

I had a spirited debate with our twitter friend, if you want to see it all, check out my twitter @lgvamos3
Don’t worry it remained civil.

Then, we were all promptly called out by Dan D Bradley from ontheforecheck. Well done Dan.

But where does Roman Josi fit on the Nashville Predators depth chart? For the most part, it is some form of: Josi flip-flopping with Subban at 1 and 2, then Ekholm switching around with Ellis for 3 and 4.

Is there any validity to the advanced stats that Roman Josi might be a lot of flash but little actual defensive impact? Is he a one-way defenseman? Can he actually be the 4th best of 6 defensemen?

There are tables like the above that show Roman Josi and Shea Weber giving up more shots than their defensive counterparts. They ranked 2nd and 3rd worst on the D-corps. How can that be? Aren’t they elite defenseman? Or are GM’s who get paid millions of dollars to evaluate players wrong on these guys? Probably not. There is more to this than a few tables.

It essentially comes down to how much faith you put into Quality of Competition.

Before we truly dive into this there is a more detailed view of Predlines breaking down the analytics of our defenseman… read Breaking Down the Defense, Part 1.

I’ll give you the short version in this article, but if you are a nerd like me, read Parts 1 and 2.

must read: Breaking Down the Preds Defense

Quality of Competition

Anyone that watched the Nashville Predators last year knows that Shea Weber and Roman Josi were like white on rice with the opposing teams forward stars. Just how attached were they?

No other pair in the NHL faced tougher competition night in and night out. Roman Josi faced the 11th stiffest competition and Shea Weber ranked 15th in the same metric. No other team that made the playoffs had a pair in the top 20 in regards to quality of competition.

But Ekholm and Ellis are a good pair, surely they also face tough competition… right? Well, not really. Ekholm was ranked 81st and Ellis was ranked 120th in terms of quality of competition among NHL D-men. Ekholm played almost exactly average competition and Ellis had about 75% of NHL defenseman playing against better/tougher competition than he did. That’s a rather large difference between the 1st and 2nd D-pair.

Many, including our twitter friend, claim Quality of Competition is tough to measure, and thus, cannot base an argument off of.

This is where the fissure between analytics and the eye-test widens.

If Quality of Competition is too imperfect to take into account, which I disagree with… then the eye-test is necessary. There is probably not a 4th liner in the league who has ever made eye contact with Shea Weber or Roman Josi. Obviously, I’m exaggerating but you get my point.

While it might not be worth debating the quality of competition difference between two players ranked 10th and 11th, or even 10th to 30th, a gap like we see between the 1st and 2nd pairs last season must be taken into account.

Roman Josi and Shea Weber together averaged the 13th toughest competition in the league. No single defensive pair averaged tougher match-ups.

Meanwhile, Ekholm and Ellis (that have great fancy stats all across the board) averaged 103rd in terms of quality of competition. Meaning, about 102 defensemen in the NHL, played tougher minutes than Ekholm and Ellis did as a pair. 13th to 103rd cannot be ignored, I don’t care how imperfect the measurement is.

Shot Selection

Yes, Roman Josi and Shea Weber for that matter do not have terrific fancy-stats across the board, but High Danger Scoring Chances or HDSC, dispute claims that the pair is “bad at defense”.

More from Predlines

If we take into account shot selection, both players were positive in HDSC. Weber was +52 and Josi was +25 at even strength.

I like that metric better than Corsi because it ignores irrelevant flips at the net. Many criticize the pairs Corsi but that ignores where the shots are coming from. A shot from the red line is treated the same as a breakaway. Defensemen are trained to let “easy” shots get to their goaltender… any attempt at blocking it might just deflect it into your own net.

As I explained here in much greater detail (with some decent evidence if I do say so myself) Roman Josi and Shea Weber were the best pair in the entire NHL last season. No other pair could play the minutes they did and come out alive.

Related Story: Breaking Down the Defense, Part 2

Saying that Roman Josi is the 4th best defenseman on the team due to some advanced metrics while ignoring other advanced metrics (like Quality of Competition and HDSC) is simply irresponsible and a symptom of tunnel-vision. No statistic exists in a vacuum.

Quality of Competition and HDSC explains and exonerates Roman Josi and Shea Weber in many “deeper statistic” shortfalls. Shea Weber has been horribly criticized after being traded for many of his advanced numbers. Hockey fans who believe that either player is bad or one-sided is too far into analytics and/or not looking at the right numbers like I have presented. Like I said earlier, there is a happy medium between the numbers and the eye-test.

Next: Preds Third Defensive Pair Will Be Interesting

Sometimes it just matters that you look at all the stats, not just the numbers that you want to show.