Edmonton Oilers

Bearing off

/ sunilagni / 1 Comment

This has to be it, right? The Edmonton Oilers get a second chance to win the Stanley Cup starting on Wednesday night against the Florida Panthers, which should be a very entertaining series. Both clubs have been dominant this post-season, and it felt kind of inevitable that they’d be going at it again. Pip count is near identical, now it comes down to the rolls – and hopefully things swing the Oilers way.

Here’s how the two clubs have done at even-strength this post-season. Both have dominated the scoresheets, each hovering around a near identical 62 percent goal-share. And their success has been supported by strong shot-share numbers – no surprise there.

What stands out here is the fact that the Oilers goaltending has been slightly better at even-strength this post-season, with the Panthers hovering closer to league average levels. Bobrovsky had a solid regular season, ranking 15th among 58 goalies who played at least 1,000 minutes in the regular season with a 91.9 percent save percentage. And a +8.53 goals-saved-above-average. Skinner, meanwhile, ranked 46th in save percentage (90.2 percent) and 48th in GSAA (-8.80). What that tells me is that Bobrovsky put together longer stretches of success in the regular season than Skinner – but it remains to be seen which one can do it first in the finals.

Both teams are also getting good results from across their rosters – scoring goals with and without their star players on the ice. Without one or both of McDavid or Draisaitl at even-strength, so about 50 percent of the team’s total time, the Oilers have posted a goal-share of 67.74 percent (21 goals for, 10 goals against). And on the other side, without one or both of Barkov or Tkachuk, about 46 percent of the team’s total time, the Panthers have also posted an exceptional goal-share of 66.67 percent (20 goals for, 10 against).

Where the Panthers depth players have an edge over Edmonton is their possession numbers, as measured by the Corsi For percentage (i.e., unblocked shot attempts). Edmonton tends to struggle controlling the overall flow of play, especially with Nurse on the ice with the depth players. In those situations (i.e., Nurse without one or both of McDavid or Draisaitl), the results have been good (9 GF/5 GA), but the team’s shot share metrics like Corsi and Expected Goals hover around 45 percent.

Here’s how the Panthers skaters have fared this post-season at even-strength, sorted by ice time, with a basic heat map applied to show how each player compares to their teammates.

While there are some players riding the PDO wave, like the Lundell/Marchand/Luostarinen line (who have been excellent), there are others like Reinhart who have the potential to see their scoring pop at evens. Reinhart’s on-ice shot-share numbers have been excellent, but he has only broken-even in terms of goals. Suspect the Panthers coaching staff will heavily target Darnell Nurse, as well as Henrique, Brown and Janmark. As I wrote recently, the Oilers tend to allow a lot of shots with these players on the ice – a trend that has continued since the regular season.

Should also note that the Panthers powerplay has only been alright. They’ve scored at a rate of 8.47 goals per hour, which is close to what they had in the regular season. They’re struggling to generate shots, sitting at a rate of 38 shots per hour, which is one of the lowest among all teams this post-season, and well below where they were in the regular season. The Panthers penalty kill on the other hand has been excellent. They’ve only allowed seven this post-season (a rate of 3.90 goals against per hour), which is one of the best in the league and half of what they allowed in the regular season. They’ve done an excellent job suppressing shots and making life easier for their goaltender – something the Oilers powerplay will have to figure out right away.

As I said at the top, both clubs have moved into their home boards and are in excellent position to win the game – but it’ll come down to the rolls. A bounce here and there, and hopefully someone getting the hot hand should allow Edmonton to bear off and win their first championship since 1990.

Data: Natural Stat Trick

Racing

/ sunilagni / 1 Comment

Couple notes from the first four games of the western conference finals. Ahead in the race to the finals, the Oilers have a massive opportunity tonight to close things off.

  • The Oilers results have been excellent at even-strength, outscoring the Stars 9-3. Shooting percentage is close to 11 percent and the team save percentage is over 96 percent – both well above normal levels for the league and for the team.
  • The Oilers underlying shot-share numbers at even-strength have been fine. While their adjusted Corsi For percentage is at 49 percent and their Expected Goals for percentage has been 47 percent, it’s largely due to the fact that they’ve been leading for about 69 percent of the total time at even-strength.
  • When the score has been within one goal (about 132 minutes of 192 total), the Oilers have posted a Corsi For percentage of just under 53 percent, and an Expected Goals For percentage of 51 percent.
  • Stuart Skinner has arguably been the best player for Edmonton. After a shaky start, and posting shorthanded numbers similar to his career levels, he’s significantly improved. Question remains if he can maintain this level, or if he fades off – similar to what’s happened in the past when he’s playing too many minutes.
  • Oilers powerplay has been great – five goals in about 20 minutes translates to a scoring rate of 14.42 goals per hour. That’s well above what they finished with in the regular season (8.90 per hour) and higher than the best powerplay (Winnipeg had 11.06 goals per hour). The Oilers rate of shots is slightly higher than where they were in the regular season, so it’s too early to panic about their powerplay falling off.
  • The penalty kill results are much better now that the goalie is making saves. Who knew.

Below is how the Oilers skaters have performed at even-strength against the Stars so far.

Couple more notes:

  • It makes a lot of sense to have Stecher take a break. He’s getting a lot of attention for stabilizing Nurse and being a “gamer”. But he’s been on the ice for a lot of shots against relative to his teammates. Nurse has also struggled, but continues to see a lot of ice time. Not sure if Nurse playing with Kulak, which is expected today with Ekholm returning, makes a lot of sense. Their on-ice shot share numbers have been poor in the regular season and in the playoffs.
  • Losing Hyman is massive. The team tends to perform well with him on the ice, regardless of who’s on the ice with him – a rare driver in the league. Kane is going to get more minutes, but I wouldn’t expect him to last long, especially if he’s playing against better competition.
  • That entire bottom six has not been good this round. While the Oilers have outscored the Stars 3-1 at even-strength without McDavid or Draisaitl, their underlying shot share numbers have been dreadful. The team’s Corsi For percentage without one or both of the glimmer twins is at 40 percent and the Expected Goals percentage is at 37 percent.
  • Adding J. Skinner to the forward group would be recommended. Or perhaps moving Nugent-Hopkins to a third line to see if he can help others. In about 15 minutes away from Draisaitl and McDavid this round, RNH’s on-ice shot share numbers have been over 55 percent – he’s been excellent this series.

Data: Natural Stat Trick

Goaltending is letting the Edmonton Oilers penalty kill down again

/ sunilagni / 1 Comment

Seeing the Edmonton Oilers get awful goaltending on the penalty kill this round should not be a surprise to anyone. Especially those that follow the team’s regular season numbers.

For two straight seasons now, the Oilers skaters have done a decent job preventing shots and chances against on the penalty kill. The skaters have done their jobs disrupting powerplays, getting in lanes and blocking shots – everything you expect them to do and hear coaches and commentators talk about.

In the 2023/24 regular season, the Oilers allowed the third lowest rate of unblocked shot attempts (67.78 per hour), the fifth lowest rate of shots on goal (49.27) and the third lowest rate of expected goals (6.90). But because their goaltending could only stop 84.68 percent of the shots, which ranked 27th in the league, the penalty kill finished 15th overall in terms of actual goals against. The tactics and process, which a team controls, were solid. The goaltending was not.

Similar things happened in this recent 2024/25 regular season. Again, the skaters allowed some of the lowest rates of shots and chances against in the league – ranking seventh best in unblocked shot attempts (71.08), eighth best in shots against (48.68), and third best in expected goals against (7.48). Despite all of this work and commitment to strong defensive play, the actual results were mediocre with the Oilers ranking 17th in terms of goals against. And this was again caused by poor goaltending, which ranked 23rd in the league (83.70 percent).

Among the 59 goalies who played at least 100 shorthanded minutes over the last two season, Stuart Skinner ranks 40th in save percentage (85.40 percent) and 43rd in goals saved above average (-1.86).

Evaluating a penalty kill is pretty straight-froward. The skaters are responsible for limiting shots and chances against. Goaltenders stop the pucks that get through. You can measure both components with different metrics, and assign accountability – it’s not complicated.

The Oilers penalty kill has been a disaster this post-season. Part of it is the fact that the Oilers are allowing slightly more shots and chances against. But the bigger issue is that in each round, the Oilers have had the some of the worst goaltending save percentages. In the first round, the Oilers team save percentage was the second worst among sixteen teams with 75.00 percent. In the second round, it was the second worst among eight teams with 82.35 percent. And it appears this trend will continue against Dallas in the western conference finals.

Hopefully the goaltending on the penalty kill can straighten out and maybe get closer to the levels it was at last post-season. But based on the goaltending numbers from the last two regular seasons and the last two rounds of this post-season, it’s hard to have confidence. It’s really going to be on the Oilers dominating at even-strength and the powerplay to outscore and mask the penalty kill’s deficiencies.

Data: Natural Stat Trick

Timing

/ sunilagni / 1 Comment

Impressive run for the Edmonton Oilers so far, reaching the western conference finals for the third time in four years. And beating two solid teams along the way, including the division champs.

What’s really stood out this post-season are the contributions from across the roster. Especially from the individuals who had been dealing with injuries, and who I thought would be non-factors. That includes Kane and Klingberg, who are coming off major, career-altering surgeries. And Frederic who appeared to be damaged goods at the trade deadline and had a shaky start in the first series against Los Angeles. Glad to see that the bets that management took have worked out so far. The timing of it all has been excellent.

After coming off a regular season where the team posted a -22 goal differential (66 goals for, 88 goals against, a 42.8 percent goal-share ) at five-on-five without McDavid or Draisaitl on the ice, things have certainly turned around. The depth outscored Los Angeles 9-7 (a 56.25 percent goal-share), and outscored Vegas 7-2 (a 77.78 goal-share). Especially impressive has been the fact that their underlying shot-share numbers have improved from the regular season.

(Just waiting on NST to update their team-level data, and I’ll add the supporting details here.)

And here’s how the individual roster players have fared at even-strength (5v5) after the first two rounds of the post-season, sorted by time on ice. We’re dealing with small sample sizes. But the one thing that stands out is the fact that the Oilers have more than one anchor on the blueline that can help drive play and scoring chances without McDavid and Draisaitl on the ice.

Bouchard continues to thrive as he did during the regular season with and without the star players. But in the post-season, we’ve seen the emergence of the Walman/Klingberg pairing that’s been making a lot of positive plays. Both defencemen dealt with injuries at the end of the regular season, so their performances in the post-season have been a pleasant surprise. And it appears their success together (GF) is sustainable. Again, it’s been great timing.

Below are the defensive pairs that the Oilers coaching staff has deployed so far at 5v5, sorted by time-on-ice (minimum 15 minutes played). Think you can feel pretty confident with Bouchard paired with anybody, but Kulak is the best fit. The only tandem the team should really avoid is Nurse-Kulak – they don’t work stylistically statistically, in my opinion. The good news is that both player have success with others, and there are options for the coaching staff depending on the situation and risk-tolerance. Nurse and Stecher are posting good numbers together, and should be the go-to third pairing.

Data: Natural Stat Trick

Remembering Brian King

/ sunilagni / Leave a comment

Sad to hear the passing of Brian King, the man behind the popular PDO metric in hockey. Brian was a very influential person in the Edmonton Oilers fan community and had a huge impact on my work and others. I’ve referenced his PDO stat thousands of times on my blog and other outlets, and it’s a wonderful metric that’ll be referenced for a long, long time.

Brian helped shape how fans view and analyze the game of hockey, especially when it comes to evaluating individual players and the ‘puck luck’ they may or may not have. The PDO metric will continue to be referenced when an objective take of the game is needed, something Brian regularly provided online. Combining his metric with other stats and insights has made the game so much more enjoyable for me, and I’m sure for others as well. I’ll always be thankful to him for that.

I had the pleasure of interacting with Brian regularly over the years. He even took the time to talk Oilers and how he got into hockey and stats on my podcast back in 2022. He talked about coming up with PDO in the summer of 2008 while talking about the Oilers decision-making with other like-minded fans on the Irreverent Oiler Fans blog (which was maintained by Vic Ferrari/Tim Barnes). The backstory of PDO is also in Bob McKenzie’s 2014 book Hockey Confidential – something I was just thrilled to see when I first read it.

So while Brian is gone, he definitely won’t be forgotten.

Deepest condolences to Brian’s family and friends, and those feeling his loss.

Direct range: Oilers heading into the playoffs in poor form

/ sunilagni / 2 Comments

The Edmonton Oilers are heading into the playoffs in not-so-great shape.

Over the final twenty-five games of the season (so since around late February), the Oilers had a record of 14-10-1, a points percentage of 0.580 that ranked 14th in the league. Their even-strength (5v5) goal-share of 43 percent (43 goals for, 57 goals against) ranked 27th in the league, only ahead of New Jersey, San Jose, Chicago, Nashville, and Minnesota.

Injuries were obviously an issue and played a role in the team’s shooting percentage, which ranked 30th during this period, and save percentage, which ranked 27th. But it’s important to note that even before this final stretch the Oilers weren’t too great in either of those categories as they were below league average pretty much all season. And because of this, the Oilers are heading into the playoffs with one of the worst goal differentials at even-strength (-3) among playoff teams.

The good news is that despite the injuries, the Oilers posted decent shot-share numbers with their Corsi For percentage and Expected Goals for percentage ranking higher up in the league. So, there’s reasons to believe that their goal-share could improve. But it’s important to note that their shot-share numbers were largely star driven, as the team could only post a 48 percent Corsi For percentage and a 46 percent Expected Goals for percentage when McDavid and Draisaitl were not on the ice. So it’s hard to have confidence in the depth players being real drivers at evens. Should note here that that I excluded the games when the glimmer twins were not in the lineup.

The other major issue heading into the playoffs is the Oilers goaltending, which has been inconsistent all season and below league average levels. And it’s where the Kings have a significant edge.

Among the 62 goalies who played at least 1000 minutes this season, or at about 20 games (all situations), Darcy Kuemper ranked fourth overall in terms of save percentage (0.921) and goals-saved above average (GSAA, +26.76), only behind Connor Hellebuyck and Andrei Vasilevskiy. Stuart Skinner on the other hand ranked 39th with a 0.897 save percentage and 42nd with a GSAA of -4.44. Among the true regular goalies (ones who played at least 50 games this year) Skinner ranks near the bottom of the list. And for those wondering, Pickard ranks 24th in terms of save percentage (in the list of 62 goalies) and 26th in GSAA.

Should note that both Los Angeles and Edmonton were very good at limiting shots and chances against at even-strength and the penalty kill this season. One team had goaltending and had a shot at the division title. The other one had Stuart Skinner. It’s a major weak spot for the roster, and a position that Oilers management chose not to address last off-season or at the trade deadline.

Data: Natural Stat Trick

Covering Blots: Digging into whether Darnell Nurse has been good enough for the Oilers this season

/ sunilagni / 1 Comment

Defenceman Darnell Nurse has had a lot of highs and lows throughout his season (and career) with the Edmonton Oilers.

Early in the campaign, he posted some of the team’s worst on-ice shot-share numbers and goal-differentials. But by early December we saw his performance and results turn around after the coaching staff played him more often with Leon Draisaitl, who has had an MVP-level season. As the year progressed, Nurse made some great plays and contributed on the score sheet. But that would again be followed up with blunders in his own zone, leading to opponents sustaining pressure and generating scoring chances. His game has recently improved, which is great considering the playoffs are coming up. But it remains to be seen if he can continue playing at this level.

This all led me to a pretty straightforward question: is Darnell Nurse having a good season?

To determine if he meets the threshold of “good”, I produced some basic criteria for a defenceman that is being paid $9 million per year:

  • Must be playing regular minutes against the other team’s top competition at even-strength (5v5), and top minutes on the penalty kill.
  • Must have a positive impact on the team’s shot and scoring chance differentials at even-strength (5v5).
  • Must have a positive impact when playing with the team’s depth players, as in when McDavid and Draisaitl are on the bench.

There could be other criteria, but what I have above are things that correlate to scoring goals and winning games. It’s basically the bare minimum.

Ice-time and Usage

Nurse is currently averaging just under 23 minutes a game this season (in all situations), which ranks third among Oilers defencemen. And he ranks first in the average number of even-strength minutes (19 minutes a game). What’s interesting is that 25 percent of his even-strength minutes this season are against elite competition, according to the methodology used by Puck IQ. This is a drop from last season when Nurse played 30 percent of his minutes against top competition.

The table below shows a steady decline in Nurse’s proportion of ice time against elite-level competition (orange bars) throughout his career, as he was over 40 percent just three seasons ago. The 25 percent proportion he has this season would be one of the lowest in his career, indicating that the coaching staff is well aware of his deficiencies, and are actively trying to put him in a position to succeed.

.

Another thing worth considering is the steady decline in Nurse’s minutes on the penalty kill. Nurse is averaging below two minutes a game this season, which is a significant drop off from the last four seasons when he was regularly on the top penalty-kill unit. The graph below shows his average ice time per season over the course of his career, and is a good indicator of the trust the coaching staff has in the player.

.

Nurse’s Impact at 5v5

Below are Nurse’s relative-to-team shot-differential numbers by season. I tend to use Corsi, a proxy for puck possession, and Fenwick and Expected goals as a proxy for scoring chances.

.

This is going to be the third straight season where the Edmonton Oilers have seen their shot and scoring chances numbers take a hit when Nurse has been on the ice.

Yes, he does play a lot of minutes and sees top competition. But as we saw above, those hard minutes are being reduced. He’s clearly not a player that can be relied on to drive play, and he’ll need to be with a strong defensive partner and high-end forwards as much as possible for him to provide value to the team at even-strength. Which leads to the next criteria.

Impact when playing with depth players

When the Oilers have Nurse on the ice with the depth players, so no McDavid or Draisaitl on the ice, the Oilers have posted some pretty poor performance numbers and results. Their Corsi For percentage and Expected Goals For percentage both drop down to 46 percent when the depth players are playing with Nurse, which, as mentioned above, is why the coaching staff was okay with getting Nurse away from them earlier in the season.

The good news is that the results have been good, with a goal differential of plus-four (goal share of 55 percent), but that’s been largely luck-driven. The team tends to play less often with the puck when Nurse is out there with depth players, so it’s hard to have confidence in these results over the long run.

Final Thoughts

While we sometimes see some great plays from Nurse and his name on the scoresheet, it’s hard to ignore his usage, performance and results over the course of the full season. Expectations are rightfully high for the player, and unfortunately, he’s not meeting some basic criteria that correlates with winning games. The hope is that he can play well into the playoffs, but it’s hard to have confidence when we look at the data over a full season.

Based on his salary and role on the team, you would expect Darnell Nurse to be one of the best defencemen in the league, though you could make the argument that he might not even be the third-best defenceman on the Oilers. That’s a problem for management to address this off-season.

Data: Natural Stat Trick, PuckIQ

Also posted at Oilersnation.

Talking sports analytics on the TSI Today podcast, hosted by Sport Central

/ sunilagni / Leave a comment

I was recently invited to talk sports analytics on the TSI Today podcast, which was hosted by Sport Central here in Edmonton. Full segment is below.

Topics we covered:

  • How I got into hockey analytics and sports writing, and some of my methods when reviewing and interpreting data.
  • Importance of hockey analytics, the role fans have had in growing it and how front offices are leveraging it.
  • My thoughts on the skillsets required to get into the sports analytics industry and recommendations for those looking to get into it.

Big thank you to Chad Tomaschefski from Tridon Communications for the interview, and Janna Tominuk from Sport Central for hosting. And special thanks to Mark Spector for the invite and the tour of Sport Central.

For more information on how you can help a great cause, check out Sport Central’s website: https://www.sportcentral.org

You can also checkout the interviews with Janna and Mark below:

.

Words, numbers, ideas – Remembering Bruce McCurdy

/ sunilagni / Leave a comment

Devastated to hear the passing of my friend Bruce McCurdy. He was an absolute gem of a person who was incredibly knowledgeable and passionate about hockey and local sports. Had the pleasure of meeting him many times over the years and getting to know him. I’ll miss our hockey and non-hockey chats and exchanges.

To me, Bruce was the unofficial historian of the Edmonton Oilers – someone who could recall intricate details about the team’s history, the players, the games. What made Bruce stand out was his ability to give the context around the team and what was happening at the time of the events. He could recall what the fans were feeling, what was being written about the team, and what else was happening in the league. He always helped paint a picture and helped you feel more connected to the Edmonton Oilers – more than what anyone else could. I think that’s why this feels like such a loss for local fans.

His ability to blend words, numbers and ideas so effectively was a huge influence on me and others, and is going to be deeply missed. He was willing to engage with others and challenge others. He brought forward well-informed ideas and opinions, and was always open to new ideas. He was a great hockey writer, making his loss even more profound. He’ll always be an iconic figure in the Oilers blogosphere.

Bruce was incredibly supportive of me and my work over the years. He was very encouraging, especially early on, and willing to share my work and give me feedback. What I appreciated most about Bruce was when he’d reach out to correct something I had written or something I might have missed. He’s a big reason why I’ve been writing for this long and willing to take risks along the way. I owe a lot of my success to Bruce.

Condolences to his family and those feeling his loss.

The Oilers have gone from red hot to a hot mess

/ sunilagni / Leave a comment

The Edmonton Oilers have been very poor since the 4 Nations Face-Off. In 10 games since the break, they have three wins and seven losses, having been outscored 27-40 in all situations.

Even-strength (5v5) has especially been awful as the club went from a +14 goal differential heading into the break to now sitting at even. Put another way, it took them 55 games to have a goal-share of 53 percent, which was eighth best in the league. And all of that progress, especially what they did in January, has been wiped out in the last ten games. They now rank 18th in the league with a goal share of 50 percent. Not exactly looking like a playoff threat. A graph showing the Oilers cumulative goal differential this season at even-strength is below.

.

The drop-off in results has been driven by their inability to control the flow and their spending less and less time with the puck. Prior to the break, the Oilers had some of the best shot-share numbers in the league, hovering around the 55 percent mark when it came to Corsi percentage (which is a proxy for puck possession) and Expected Goals (which is a proxy for scoring chances). But since the break, these numbers have fallen to league average levels, with their Expected Goals percentage even dropping below 50 percent.

For context, below is the Oilers underlying shot-share metrics in rolling 10-game segments. I’ve added a marker at the 55 game mark, which is the first game after the Four Nations Faceoff tournament. The other low point came around the game 34 mark, which was in late December when the team was struggling. In January, the team turned things around and started playing closer to what was expected. But things appear to have fallen off again. The recent numbers are well below where they were prior to the break and an indication that the tactics and process the coaching staff has in place for even-strength play isn’t working. The results are suffering because of it, and it has to be addressed as soon as possible.

.

There are two issues to consider when trying to understand why these numbers are falling off.

The first is that the team’s goaltending has been a weak spot for the team for a long time now, and hasn’t at any point this season been at or above league average levels. Because of this weakness in net, the team seems to be playing a little too safe now and not pushing for offence as frequently. Instead of taking chances and really leveraging the offensive talent they have up front, there are a few too many conservative plays happening, especially in this recent stretch of games.

The switch to playing more conservatively at even-strength is reflected in the team’s declining rate of generating shot attempts. Before the tournament break, the Oilers were generating 62.6 shot attempts per hour — one of the highest in the league. In the last ten games, that rate has dropped by 8.8 percent to 57.04 shot attempts per hour — which is slightly below league average.  Worth mentioning that the team’s rate of shot attempts against has remained steady all season.

.

And as I wrote about a month ago, the team has also been getting a lot more shots from their defencemen, which isn’t ideal considering the probability of scoring drops off the further you’re away from the net. There appears to be more hesitancy to make plays and take some risks to get the puck moving into high danger scoring areas. Instead, the team is taking shot attempts from distance, and hoping the best for rebounds. This is a weird tactical issue that various coaching staffs have tried in Edmonton, and it just doesn’t work well.

Considering the team’s championship aspirations, and the importance of home-ice, it’s critical that the Oilers coaching staff recognize the issues and make the necessary adjustments. The chase for a division title is slowly slipping away, but there is still time to make some ground.

Data: Natural Stat Trick

Also posted at Oilersnation.