He shoots, he scores!!!
Except most of the time, he doesn’t.
Take Mattias Ekholm for example. During 5v5 play this past year, Ekholm took a shot 230 times and scored only three goals. That’s a conversion rate of 1%. Then there’s Filip Forsberg, who shot 234 times this season and scored 31 goals, a conversion rate of 13%. Both players shot roughly the same number of times, but had very different results.
In this series, I’ve been looking at the three ways that a shot can go wrong (blocks, misses, and saves) while also looking at which players on the Nashville Predators are better at avoiding those obstacles to goal-scoring.
In part 1, we saw that the Predators excelled at taking shots but were entirely average at scoring. In part 2, we saw that Craig Smith was the second best Predator at taking shots, but the worst at scoring. Viktor Arvidsson and Kevin Fiala on the other hand were studs at both.
Today, I will look at the Predators' shooting again, but this time within the context of the entire league. You can’t really decide whether Fiala was all that great unless you know how he compares to the entire NHL. Was he great or was he just great on the Preds? Was Smith really “Smithing it”, or does he just look bad on a team of finishers? And what about the new free agent acquisitions, Nick Bonino and Scott Hartnell?
To make cross-team comparisons valid, this is only going to look at regular season 5v5 shots, and it will exclude any player with less than 500 minutes of TOI, who shot less than 50 times, or who played for multiple teams. If you don’t do this, you end up saying players like McLeod were your best shooters. This takes us down to 499 players.
Ready? Let’s do this.
First, we look at Corsi (shot attempts) per 60 minutes of ice time. Who excels on the team at producing shots on goal? (The lower the number on the left, the higher their rank in the league is on this graph.)
Remarkably, the Preds have five players in the top 50. Arvidsson, Smith, Forsberg, Fiala, and Neal.
Well, at least they did have five. Now they have four.
Having five is unusual. A great teams like Chicago has two, and Pittsburgh only has one. So why didn't we destroy Pittsburgh in Corsi? (We were basically even.) Well, there's a giant gap of over 100 spots between our fifth best and sixth best players (Neal to Josi). The Pens don't have such a big gap.
When I first noticed this, I thought, “Oh, that could be a problem; we don't have strong secondary shooting.” However, the reason we have such a gap is our top 5 are so good. The Pens top 5 rank lower, removing the gap. Not sure there’s a problem here.
On the back end, if it seemed like Mike Ribeiro hardly ever shot… well, it’s because he almost never did. He comes in at 489 out of 499 on our list. Only 10 people (who had over 500 minutes of TOI and 50 Corsi) shot less than him. Colton Sissons also deserves a few “shoot the puck!” yells, particularly when you look at his stats below.
Turning now to where those shot attempts can go wrong, we look at blocks, misses and stops. Blocks first.
Mike Fisher, Arvidsson, and Ryan Johansen are our top 100 guys in percentage of shot attempts blocked. In comparison, Chicago and Pittsburgh also had three, though each had a guy in the top 20 in the league. So the Preds are kind of average by this metric. Defensemen are near the back end of the ranking. with the worst six guys all being blueliners. Their overall rankings within the league may not be that bad, though, and we can’t tell that until we look at the defense separately from the forwards. (These posts can go on forever because there’s always another thing to look at.)
How about unblocked shots that miss?
The Preds show up very solidly again here with four players in the top 50—okay, 51—in the league. The players showing up there are an entirely different set from those in the top 50 for Corsi per 60, except for Fiala who scores highly on both.
Of those with very poor rankings, the Prince himself comes in 456th out of 499. I knew he came in last among Preds forwards from the previous post, but this number is baaaad. Smith shows up as the second worst forward.
Now, we look at the proportion of shots which were not stopped by the goalie.
The Preds have three in the top 100 with Sissons, Forsberg, and Ellis leading the way. This is quite comparable with Chicago with four in the top 100.
However, Pittsburgh… Pittsburgh has five in the top 100, two of whom are in the top ten. Ranking third, ninth, and 26th are a few fellows named Jake Guentzel, Evgeni Malkin, and Sidney Crosby. (Recall this is regular season only, so it does not include Guentzel’s stunning playoff goal run.)
For the Predators, defensemen bring up the rear with the giant exception of Ellis, who is third on our team and 81st in the league. I haven’t broken defense out yet, but I wouldn’t be surprised if he is one of the tops for defense in the NHL on this stat.
Wrapping this up, we might want a general number that sums up blocks, misses, and stops to generally measure success at getting shot attempts in goal. That’s the TCB number, Taking Care of Business. I’m almost certainly going to change this, but for now it’s a weighted average of the three filters on shots going in.
The Preds have three in the top 100, but only 1 below 50, and it’s not particularly low within the top 50. Moreover, that person is Sissons, who hardly ever shot (see his Corsi p60 in Figure 1), so Nashville didn’t get to take much advantage of this score for Sissons. The overall picture of TCB for the Predators is... “meh”, though Fiala and Arvidsson are doing well. Chicago is also “meh,” with their highest ranked skater in the 60s.
Pittsburgh, on the other hand:
That’s seven players in the top 100, four in the top 50, and two in the top five in the NHL. Those TCB numbers help explain Games 1 and 2 of the Stanley Cup Final (along with that offsides call and some save %s).
So now the big question: does the Preds’ TCB get any better with the additions this offseason?
(Remember there's 499 players, so being ranked 250th is average.)
Bonino and Hartnell, while neither shot at an exceptional rate, would rank right near the top in TCB score for the Preds. Only a single player, Sissons, would have been ranked higher. This bodes well for playing in Laviolette’s system. However, their strong TCB rankings are mostly based upon not missing, and — this is the topic of the next post — it’s not clear that’s a repeatable stat. Still, the scores for the Preds’ two new forwards suggest they could be solid contributors for our team. (Whether we need more than that, I’ll let you all discuss in the comments.)
I will start to give an answer to “skill vs. luck” question in the next post. It’s full of fun stuff like probability distributions and regression analysis. So stay tuned.