Who are the Oilers’ drivers on the penalty kill?

It’s been well-established that the Edmonton Oilers’ penalty kill needs to be better this upcoming season. It finished 21st in the league last season allowing a rate of 8.52 goals against per hour.

Their results were driven by their inability to limit shots and chances against and goaltending that was below league average. The Oilers penalty kill doesn’t have to be perfect and be the top-ranked group in the league. It just can’t cost them wins like it did in the early part of last season.

A goals-against rate around league average would be considered an improvement. And to achieve that, they must do a consistent job of preventing shots and chances. And hope that their goaltending remains steady.

Last week, I looked at how the penalty kill performed over the course of last season and found that while it did have very poor results to start the season, it did gradually improve. Their rate of shots against decreased by 18% in the second half of the season, getting much closer to league-average rates and the goaltending also improved. These factors helped reduce the rate of goals-against by hal

The next issue worth exploring before the start of the regular season is the player deployment from last season – and which skaters helped improve the odds of the penalty kill being successful. To do this, I looked at the on-ice numbers for each Oilers player who played at least 10 minutes on the penalty kill last season. Specifically, I wanted to know how well the team did at suppressing shots (shots against per hour, SA/60) and chances (unblocked shot attempts against per hour, FA/60) with different players on the ice. These are numbers that skaters can help influence, as they are trying to block shots, cut off passing lanes, and force the puck into low-probability scoring areas. For context, I also included in the tables each skater’s on-ice rate of goals against (i.e., actual results, GA/60), which skaters tend to have less of an influence on as goaltending plays a larger factor.

I’ve split the skaters between forwards and defencemen and sorted them by total ice time on the penalty kill. I’ve also applied a basic heat map to each metric to show how each player compares to the rest of their team. Green is good. Red is not so good.

The first thing that stands out is the poor shot-suppression numbers with Nugent-Hopkins, Nurse or Ceci on the ice. At least one of them was killing penalties for 72% of the team’s total ice time last season and were clearly Woodcroft’s go-to players. And since they’re on the first unit, it’s expected that their rate of shots and chances is higher relative to their teammates as they’re facing the other team’s top units and typically their top offensive players. But even relative to the other units in the league, their on-ice numbers were pretty high.

On average, first-unit penalty kills allow 57.78 shots against per hour, which is around 5 shots more than their team’s second unit (about a 7% increase) as they’re more often up against the opposing team’s top powerplay units and their star players. The Oilers’ top unit allowed 69.26 shots against per hour, which is around 20% higher than the league average. Please note, that the on-ice numbers for the skater with the highest total ice time for their team were used as a proxy for their team’s top penalty kill unit.

The top unit needs some changes, and hopefully, someone like Ekholm can be part of the solution. He was excellent for the team in the 45 minutes he played on the penalty kill last season, clearly outperforming Barrie, and has plenty of experience shorthanded. Perhaps pairing him with Broberg who showed well in limited minutes last season could help drive down the rate of shots and chances against. Desharnais was also a big reason why the Oilers’ rate of shots against improved over the course of last season. He’s a depth player well-suited to kill penalties.

Up front, someone like Ryan McLeod could likely take on more minutes shorthanded as he played over 90 minutes last season and posted solid shot-suppression numbers relative to his teammates. Foegele is another player who posted decent shot-share numbers, but his actual results may be what holds him back from more minutes from the coaching staff. And I’m curious to see how often Connor Brown is deployed next season. He has a history of playing on first-unit penalty kills, even though his numbers weren’t that strong. This past summer, Ken Holland mentioned Brown as a penalty kill option, but he may need some time getting up to speed as he only played four games last season. Someone does need to replace Yamamoto, who posted solid shot-suppression numbers last season. And if Brown isn’t ready, I do wonder if Holloway could find a role here. So far in pre-season, he’s played the third most minutes among forwards behind Derek Ryan and Adam Erne.

Plenty of decisions for the coaching staff to make, and how they deploy players on the penalty kill will be something to monitor. As mentioned above, if they can even post league-average rates of shots and chances, and get decent goaltending, the penalty kill should significantly improve from last season.

Data: Natural Stat Trick