Sustainability

I was just reading Katya's article Maple Leafs 2017 first quarter report: Shooting their way to glory (love the title). I would like to add a couple brief comments based on the vizzes I was playing around with yesterday. I went back to 2007-08 and tried to put the Leafs first quarter in some sort of perspective.

The first thing that hit me as I explored the data from Manny Elk's amazing corsica.hockey was the Leafs shooting percentage. I mean, we know it's high but exactly how high? The viz below shows every team's shooting % 5-on-5 over the last ten years. (I excluded Vegas. The redder the square, the hotter the shooting %, bluer squares get progressively colder below the historical average of 7.84.)

The Leafs shooting % after 21 games is 11.25%, higher than any team in the last ten years. It is so high it makes the hot shooting of the Lightning, Jets, and Islanders orange by comparison. This is what smaller slices of a season can do - exaggerate the variability of some stat. The Leafs had another unbelievable season in the 2012-13 lockout season with a sh% of 10.66 over 48 games. A rare seasonal shooting % would be two standard deviations above the average at 9.6%, if the Leafs sustained the 11.25% rate they would be well past the 3 SD point and entering historical record books.

I went in this morning and looked at individual 5 on 5 shooting rates for the top forwards. It's a mixed bag but Brown, Kadri, and Auston have gone well past last year's sniper rates. Auston has that new sneakier snapshot, Kadri is on his game and I have no idea how Brown is potting all those goals while he bounces around on different lines. On the other hand, it's hard to imagine that Nylander, Marner, and Komarov won't be improving (well maybe not so much Komarov).

So is the shooting rate sustainable as a team? I highly doubt it, the odds are high that it regresses to some more 'normal' level over the full season. Last year the Leafs with a similar lineup had a Sh% of 8.3. But let's take a look at xGF/60. If it's sustainable, the quality of shot locations will probably rank very high as well.



And the Leafs' xGF/60 is very high at 2.76, not blood red like Columbus or the 2014-15 Islanders, but definitely one of the highest in recent years. Poor Edmonton and Montreal have similar xGF/60 stats but just haven't been able to fill the net. Based on this, the Leafs generation of scoring chances looks pretty sustainable: the xGF/60 is nearly the same as the 2.68 recorded last year.

So we have a team with a very high xGF/60 that looks sustainable statistically and is supported by the eye test. Babs' system and the Leafs top 9 certainly appear to place an emphasis on getting into high danger areas and our defencemen aren't known for taking a ton of shots.

Nonetheless, teams with high xGF/60 rates have never had a shooting percentage like the Leafs are having now. So I believe that some regression has to be expected to a rate closer to that of last year's 8.3%.

We've seen that anything can happen in the first quarter of a season. Leafs have banked 14 wins and 28 points, supported by a historic shooting percentage. But PDO is like gravity, eventually high-flying objects fall towards the core. Let's hope that for the Leafs the forward core is more influential than gravity.