What’s Going On With The Leafs Powerplay?

The Toronto Maple Leafs powerplay was confounding in Mike Babcock’s first season as head coach of the team.  They were 3rd in the NHL in shot attempt rate at 5v4, at 103.6 attempts per 60 minutes.  Even more impressive, they were 1st in scoring chance rate, at 28.7 scoring chances per 60 minutes, a good ways ahead of 2nd place San Jose, who fired 26.3 SCF/60.  And yet, despite all those opportunities, the Leafs struggled to score powerplay goals.  They finished 29th in the NHL in powerplay success rate and were 3rd last in S%.  

This year the story is markedly different.  The team has fallen to 23rd in the league in shot attempts, generating 84.5 CF/60, an 18% decrease in shot volume.  Scoring chances paint a slightly rosier picture.  The Leafs SC/60 at 5v4 is 11th in the league, with 24.3 SC/60 representing a drop of 15%.  But the Leafs are much more successful on the powerplay this year, with a 9th ranked powerplay that’s scoring on nearly 2% more of its shots.

What explains the difference between the Leafs powerplay in these two seasons?

ICE TIME DISTRIBUTION

One of the things that Mike Babcock has been known for is playing four forwards and one defenceman on the powerplay, rather than three forwards and two defenceman as is typical.  According to James Mirtle, Babcock “uses four forwards on the 5-on-4 unit more than any other coach in the league.”  And last year the Leafs did in fact use four forwards quite a bit on the powerplay.

In a typical 3F2D set-up, you’d expect to see 60% of the ice time going to forwards.  Last year Leafs forwards played 75% of the possible PP minutes.  That shows that the Leafs played 4F1D the vast majority of the time, as the split would be 80/20 if they did it all the time.

This year the usage of 4F1D has slipped, as forwards are receiving 69% of the possible PP minutes.  The team still uses the four forward set-up fairly regularly, but not nearly as often as last year.  And we might expect a team that plays fewer forwards on the powerplay to generate fewer shots and scoring chances, so this seems like at least a partial explanation for the change.

PERSONNEL – FORWARDS

The next thing we’ll take a look at is personnel.  First, let’s compare the forwards who’ve played on the powerplay in each of the last two seasons to see if their results have changed.

pp_forwards

Two things jump out at me here.  The first is that the rate of shot attempts taken by each player (“individual Corsi for”) has fallen pretty dramatically for everyone other than William Nylander.  Nazem Kadri and James van Riemsdyk, in particular, have gone from firing a very high rate of shots to much lower levels.

The other thing that I find quite interesting is that the average shot distance for these players has also fallen a fair bit.  So while they’re taking fewer attempts overall, the ones they are taking are coming from closer to the net.  Those are shots that we’d expect to see go into the net at a higher rate, so are the Leafs taking fewer but better shots?  It certainly seems possible.  You can see that the rate of scoring chances for these players doesn’t follow the same trajectory as the rate of shot attempts, which does support the theory that they’re just taking better shots.

But what about the players who are different between last year and this year?  There are four forwards who got a lot of powerplay minutes for the Leafs last year and aren’t getting them this year: Brad Boyes, PA Parenteau, Peter Holland, and Joffrey Lupul.  Boyes, Lupul, and Parenteau all fired 22-24 shot attempts per 60 minutes at 5v4 last year, quite a high rate.  But the scoring chance data is where the loss of some key personnel becomes most clear:

Joffrey Lupul had an impressive 13.1 scoring chances per 60 minutes on the powerplay last year, 2nd on the team after JVR.  That result isn’t just a one-off either, as Lupul has consistently been near the top of the Leafs in terms of generating powerplay chances.  While Lupul’s injury problems have clearly slowed him down in recent years, he’s remained a potent powerplay performer, and losing him has likely slowed down the rate at which the Leafs are able to generate opportunities on the powerplay.  Taking Peter Holland off the powerplay may be slowing it down as well.

But what about the players they’ve added?  There are two new players on the Leafs powerplay this season: Auston Matthews and Mitch Marner.  Both players are generating shot attempts at a decent rate, but neither is generating scoring chances as consistently as the players they’ve replaced.  Matthews, in particular, is really struggling to generate scoring chances on the powerplay, as his 2.4 scoring chances per 60 minutes at 5v4 is rather low (even Leo Komarov generates nearly twice as many).  Matthews skill with the puck is obvious, so I’m not worried that his struggles will continue in the long-term, but in the short-term they are part of the reason the powerplay is generating fewer chances.

PERSONNEL – DEFENCE

The defencemen the Leafs use on the powerplay has changed a fair deal from last season.  Only two players – Morgan Rielly and Jake Gardiner – featured on last year’s powerplay and remain there this season.  Let’s see how their numbers compare between the two seasons:

pp_defence

It’s a very different story than we saw with the forwards.  Both players are shooting nearly identically to how they did the previous season.

What about changes in personnel.  Last year the Leafs gave a lot of powerplay minutes to Dion Phaneuf, who takes quite a high rate of shots.  His 31.7 iCF/60 ranked 2nd highest on the Leafs last year, with only Nylander shooting more often.  His rate of scoring chance generation was also higher than any Leafs defenceman.  So losing Phaneuf does seem to have hurt the powerplay a bit.

The new defencemen to get PP time this season, Nikita Zaitsev and Connor Carrick, are failing to generate opportunities at the rate that either Gardiner or Rielly does, let alone the rate Phaneuf did, so that’s also a partial explanation for why the Leafs are generating less offence at 5v4 this season.

CONCLUSION

There’s no simple single reason for the slowdown in the rate at which the Leafs have created shots and chances on the powerplay this season.  Changing personnel – in particular losing Dion Phaneuf and Joffrey Lupul, two powerplay specialists – seems to have played a role.  So too does the increasing amoung of powerplay time given to defencemen, especially as Zaitsev and Carrick don’t generate as much offence as the forwards they’re replacing would.

The most interesting change, though, is the possibility that the Leafs are taking fewer, better shots.  The forwards are pretty uniformly shooting from closer to the net this season than they were last season, and those who have stuck around have seen their rate of scoring chances remain fairly constant despite their raw shot attempt totals dropping.  This does suggest the Leafs forwards may be holding on to the puck for better chances, though further study would be necessary to confirm if that’s really what’s going on.

  • Ben

    Nice article. I agree with the idea that the nature of the shooters on the power play has changed, and that those changes are the main driver behind the change in success and shot rates for the Leafs.

    Also, shot attempts seems to hold more value as an even strength measure, particularly shot attempts for. You’re using a bigger sample and studying a more ‘normal’ game state. Shooter quality and high danger scoring chance generation play more into the PP.

    Long story short I don’t imagine the correlation between PP shot attempts and success rate would be as strong as, say, 5v5 Corsi close and win rates.

  • tealeaves

    I wouldn’t underestimate the negative impact martin is having on the PP. In fact, based on my analytics research and from smart people I follow on twitter, he is the primary reason for everything wrong with the leafs.

    • Stan Smith

      Maybe they need to move him to which ever side of the bench is the furthest away from the offensive zone during powerplays. If he is at the side nearest it is probably hurting the PP.

  • Stan Smith

    Going to go old school on this but if they had more shots and scoring chances, but fewer goals, last season, and this season they have fewer shots and fewer scoring chances, but scoring more goals, I would hazard a guess the reason is because the puck is going into the net more.