Optimizing the powerplay

oscar-klefbom-connor-mcdavid

One of the most interesting things about this Oilers season has been the powerplay (5v4). It’s a critical area for the team, as they finished dead last in 2017/18 and the team tried to address it in the summer by revamping the coaching staff. The team isn’t exactly a powerhouse at even-strength due to the lack of talent and depth (especially on the wings) – so it’s going to be important for the powerplay to consistently produce if they want to contend for a playoff spot.

The good news so far is that the team currently ranks 12th in the league when it comes to the rate of goals for per hour (GF/60) with 7.81, scoring 12 times. The fact that McDavid has nine powerplay points (5 goals, 4 assists – three of which were primary assists) should tell you all you need to know about who has been the offensive driver. Draisaitl and Nugent-Hopkins, who each have seven points, along with Klefbom are the go-to unit with McDavid playing around 65 of the Oilers total 92 powerplay minutes this season. The roughly 25% of the remaining powerplay time has been left for Ryan Strome and really whoever else the coaching staff feels like throwing out there.

The team as a whole isn’t doing a great job of generating unblocked shot attempts (69.67 FF/60, 19th in the league), relying more on a team shooting percentage that’s currently at 16.0%. The average team shooting percentage over the last three seasons has been 12.58, so we’ll see how long the Oilers can convert at the rate that they’re currently at.

It’s really when the first powerplay unit is on the ice that the team gets going, generating over 80 unblocked shot attempts per hour – a rate that would have them among the league’s best. And they’ve scored 10 of the teams 12 powerplay goals –  a rate of 8.84 goals per hour – which isn’t bad but really should be better considering the talent that they have.

When the second powerplay unit is deployed, it’s been dreadful – generating 39.51 unblocked shot attempts and only scoring twice (a rate of 4.94 goals per hour). It’s easy to tell why the coaching staff is giving close to 75% of the total powerplay time to the first unit.

Couple things that I think the team needs to address.

First off, the first powerplay unit is heavily relying on Nugent-Hopkins, who is also getting regular minutes on the penalty kill (4v5) and also Draisaitl, who is gradually getting more reps shorthanded. Knowing what we know about recovery time, sport science, and the importance of rest for professional athletes, I’m not sure if it’s a good idea to have your best players spending so much of their energy on both special team units. Both RNH and Draisaitl are probably young enough to handle it, but it’d probably be in the team’s best interest if the workload was distributed.

It’d really help if the coaching staff had a stronger second powerplay unit, not only to take the load off the first powerplay unit, but to also have fresher legs on the penalty kill. The team really is lacking a special teams specialist on their fourth line, similar to what they had in Letestu who at one point was really effective on the powerplay. I don’t think these commodities are hard to find; some simple research and scouting can probably uncover these.

I think a big reason why the Oilers first powerplay unit is able to consistently generate shots and chances is because they’re working the puck down low to set up the forwards. And of course it makes total sense – the talent is there up front so get the pucks to them and get shots from high probability locations. What I’ve found is that the second powerplay unit isn’t nearly as effective at getting shot from the more high danger areas – instead relying more on shot attempts form the defencemen at the point. The concept is a pretty dated one – work the puck up top, get a shot from the defenceman and look for rebounds. Unfortunately, that’s not really making the best use of your time as your chances of scoring goals are much higher when the puck is in close coming from players with scoring talent.

Using Klefbom as our proxy for the top powerplay unit, the Oilers have generated 109 shot attempts (i.e., Corsi). And of those 109 shot attempts, 18 have come from the stick of Klefbom – coming out to 16.5%. When Klefbom isn’t on the ice, the team has generated 36 shot attempts, with 15 of those coming from the stick of one of the other defencemen – a proportion of 41.7%. Ideally, the Oilers would have enough talent up front on their second powerplay unit so they’re not taking this high of a proportion of low probability shot attempts. In my mind, the team needs to fix their overall powerplay tactics and find a way to balance their first and second units.

One last thing: the drop in the rate of unblocked shot attempts from the first powerplay unit to the second powerplay unit is massive, a difference of 40.96. I took a quick look at last year’s data, comparing each team’s first and second powerplay units (using each team’s top forward in terms of ice time as a proxy) and their rate of unblocked shot attempts. On average, a team’s rate of unblocked shot attempts dropped by 18.35 when their second powerplay unit was deployed. Only one team had a drop of more than 40 unblocked shot attempts and that was the Sabres who finished 24th in the league in terms of goals for per hour.

While the team is doing alright generating goals on the powerplay, I think it’s imperative that they put a critical lens on things and find ways to sustain their success. I don’t think what they’re doing right now is sustainable, and there are significant ramifications to the team’s overall success if they don’t make adjustments.

Data: Natural Stat Trick

Advertisement

One thought on “Optimizing the powerplay

  1. Pingback: Hitchcock’s history with special teams | The SuperFan

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s