clock menu more-arrow no yes mobile

Filed under:

2013-14 Lightning advanced statistics through game 5

The season's first fancy stats report.

A Corsi event. Also pictured, a Fenwick event.
A Corsi event. Also pictured, a Fenwick event.
Joel Auerbach

The statistics I'm using today are taken from unless otherwise indicated and are for 5v5 only. As the season wears on and more games are played, we'll start looking at 5v5 close data, but for now we need to maximize the sample size.

1. PDO:

I always start off with PDO for a couple of reasons: it's one of the few short-term stats we use and it's a pretty good indicator of whether alarm is warranted or not. PDO adds shooting percentage (Goals For/Shots For) to save percentage (Saves/Shots Against). These two stats are essentially measures of shooting "luck" (which isn't really luck but randomness). If you or your team is getting a lot of luck at either end of the ice, PDO reflects it pretty quickly. Anything above 100 will come down over time. Anything below 100 will go up over time. The further away from 100 the more likely we are to see change. [Note it is also expressed as 1.000 or 1000, so don't worry about where the dot is. Extraskater uses 100.0; Behind the Net uses 1000; and HockeyAnalytics uses 1.000. They're all the same thing.]

Right now, the Lightning's team PDO is a relatively calm 104.7, good for 6th in the league. I say relatively calm because last season at this point it was 109.5. In other words, it's still high, but it isn't as bad as it could be.

What does this mean going forward? Basically the Lightning's PDO is high because shooting percentage is slightly elevated (9.9%) and even strength save percentage is very high (94.8%). The team ought to expect to score at about an 8-9% clip and save about 91 or 92 % of even-strength shots. ESSV% was even higher before the Pittsburgh game, so some regression has already started.

Individual PDOs, which can show a lot more variance than team PDO range from Mark Barberio's 125.0 to Richard Panik's 91.1. Only Tyler Johnson (97.6), Andrej Sustr (96.0), and Panik are below 100, so randomness is benefiting a number of players. Again, for perspective, at this point last season we were looking at a range of 102.1 to 125.0, with nine players above 110. Right now there are three players above 110. What we're seeing at this point is much more sustainable than last season's hot start.

There will be some regression towards 100 as players see their on-ice shooting and save percentages fall. Especially due for regression are Barberio (whose figure is due to the fact that he's only played in one game); Nate Thompson, who had never topped 970 before last season's 100.2 and who sits at 111.1 currently; and Richard Panik, who has seen on-ice shooting percentage of 7.1% and an on-ice save percentage of 84.0, far lower than any other player. I expect there to be some improvement in his fortunes, but I can't say how much.

2. Possession

There's been some wild fluctuations in the Lightning's possession play this season, from the horrible game against Chicago to the domination of the game against Florida. It's hard to know with any certainty how the team will do over the course of the season, but in general, things are looking up from last year. They are taking 48.9% of all shot attempts at 5v5 and 46.3% of all unblocked shot attempts. When the score is close (that is, when no one's trying to overcome a deficit of 2 or more goals), the Lightning are taking 49.8% of all attempts and 47.3% of unblocked attempts.

Ideally, you want those numbers to be above 50%, but this is a clear improvement from last year and it bolsters the idea that the results the Lightning are seeing are built on sounder fundamentals than we saw last season. The key will be making sure that games like the Chicago game don't happen again.

Which brings me to something that has been a real problem:

3. Special Teams

Special teams (power play and penalty kill) have been having an outsized effect on outcomes, but this has come in a weird way. Until the Pittsburgh game, the Lightning had a Special Teams Efficiency of 105.1 (PP% + PK%; it often ends up close to 100.0, like PDO). At that point, the power play was 7th overall at 28.6% and the penalty kill was 22nd overall at 76.5%. After that game the power play rate went up to 31.6% (4th) and the penalty kill dropped to 68.2% (26th).

That brings the STE to 99.8%, which indicates that the effect of the special teams as a whole are less dramatic,. However, that's basically because the horribleness of the  penalty kill is offsetting the outrageous success of the power play. Now it's up to the team to try to keep the power play going strong while fixing the penalty kill.

And the abysmal penalty kill numbers leads us to:

4. Goaltending

(no, really, it does)

Overall the Lightning have a team save percentage of .896. Same-old, same-old. They've given up 2.8 goals per game (18th in the NHL).

But at even strength, the Lightning have a save percentage of .948. Big difference. And this has affected Anders Lindback more than Ben Bishop. Bishop's overall save percentage is .928, while his ESSV% is .968, a difference of .04. Lindback's overall save percentage is .843, but his ESSV% is .917, a difference of .074.

Basically, the penalty kill has done a great deal of damage to the team's fortunes. So, you know, fix that.

In other goaltending news, two of the five games so far qualify as Quality Starts (one if the league median save percentage is higher than .912). Both of those are Bishop games, giving him a 66% QS rate, which is an elite rate. I expect that to moderate as more games are played, but Bishop's performance has certainly been helpful thus far this season.

I want to reiterate that this is a very small sample, and these numbers will certainly change as more games are played. It's far too early to start drawing firm conclusions about the team, but the numbers do back up the impression that this season is better than last. So far.