The arrival of playoff hockey also signals prime time for hot takes in the hockey world.
Some individuals will be heralded as “clutch players” forever because 30% of their shots went in the net in a small sample size, while others will be scrutinized for the remainder of their careers for various reasons, some legitimate and some not so legitimate. Alex Ovechkin could lead the playoffs in scoring, but if the Captials don’t win the Stanley Cup you know he’ll take the brunt of the blame from the likes of Don Cherry.
With this in mind, let’s take a look at some prime playoff breakout candidates as well as some prime bust candidates in terms of scoring potential.
- Patrice Bergeron (BOS) – While Bergeron’s point production has never been his biggest strength, he’s still always put up good numbers. He’s a player who’s true value has been brought to light by the advancement of hockey statistics, but he deserved more goals than he got this season. That much is apparent when you see he only converted on 7% of his shots this season, while his career shooting percentage is 9.9. Combine that with the fact that Bergeron led the NHL (among players who played >500 minutes) in 5v5 shots per hour this season, but is only converting on 5.19% of those shots and it seems as though a little positive regression in the near future is inevitable. His Bruins are also playing the Senators in the first round, which I expect them to win rather easily. Dawson Sprigings’ (@DTMAboutHeart) expected goals model doesn’t seem to think Bergeron has been as unlucky as I expected, though, putting him at 14.39 ixG at 5v5 while his actual 5v5 goal total was 12.
- Brendan Gallagher (MON) – For virtually the exact same reasons as Bergeron, I expect Gallagher to start filling the net in these playoffs. He ranks third in shots per hour at 5v5 and is only converting on 5.56% of those shots. They have to start going in the net, he’s a career 9.2% shooter at all situations who is currently converting at a rate of 5.3%. Once again, though, Sprigings’ expected goals model disagrees with my assessment, putting him at only 9.14 ixG as opposed to his actual 5v5 goal total of nine.
- Corey Perry (ANA) – Perry played in all 82 games for the Ducks this season and only scored 19 goals. This is very unlike him, as he’s always been a talented goal scorer who is an efficient shooter (career 13.2% conversion rate.) As it turns out, it seems Perry is fine, but he isn’t getting the bounces this season. He’s only shooting 8.8% at all situations which is a significant dip from his career average. This time 5v5 ixG agrees with me, as it has him at 20.33 ixG while his actual observed iG total was only 13. Perry seems like a prime candidate to break out in the playoffs, unfortunately for the Flames.
- Jordan Eberle (EDM) – Eberle, like Perry, has always been a pretty efficient shooter. He owns a career shooting percentage of 13.4%, but only converted at a rate of 9.6% this season. Eberle still managed to hit the 20-goal plateau due to a hat-trick in the final game of the regular season, but he may just be getting started. That hat-trick seems to be a sign of things to come, as his 16.34 ixG was significantly higher than his actual 5v5 goal total of ten. I expect Eberle to make some noise against the Sharks.
- Scott Wilson (?) (PIT) – I’ll be totally honest here, I have no idea who this man is, but it looks like he’s played all year for the Penguins and he deserves more than he’s gotten to show for his efforts. Wilson ranks 43rd in 5v5 shots per hour, but only has eight even strength goals to show for it. A conversion rate of 6.6% will do that to you. I assumed this guy is just a very bad finisher, until I saw him ranked 31st in Sprigings’ individual expected goals per hour. He’s also scored at a good rate in the AHL, so maybe the Penguins have another Sheary/Guentzel on their hands.
- T.J. Oshie (WSH) – Oshie is the obvious one and the inspiration for this piece. He’s had a career year with 33 goals in 68 games, mostly due to his 23.1%(!) conversion rate in all situations and his 26.14%(!!!!!!) conversion rate at 5v5. Oshie is an unrestricted free agent this summer and somebody is going to pay him an outrageous amount of money and be very disappointed on the return. He’s always been a sniper, he’s converted on 13.4% of his shots in his career, but this is absolutely outrageous. His 13.5 5v5 ixG is a testament to this, as his actual 5v5 goal total is 23. Insane. I think/hope his phenomenal luck discontinues against the Leafs in the first round.
- Rickard Rakell (ANA) – This is another gimme. Rakell exploded this season with 33 goals in 71 games, which he owes a lot to his 18.6% conversion rate. Rakell is a good player, but this is an anomaly. Rakell outperformed his 5v5 expected goal total by even more than Oshie did, potting 25 even strength goals where he could have been expected to score closer to 14.
- Michael Grabner (NYR) – Grabner began regressing a few months ago if I remember correctly, which was the most obvious bet of the season. Grabner started out on fire and at one point I think he was leading the league in goals, but he was shooting between 25 and 30 percent while sporting a career 12.5% conversion rate. Now he’s back down to 16.7%, but he still well over performed what you could reasonably expect. He still scored 20 goals at evens while he should’ve scored 12. Also, Carey Price is in the opposing net in the first round.
- Paul Byron (MTL) – Another surprising breakout year, another shooting percentage explosion. Byron enjoyed a 5v5 shooting percentage (23.75%) lower than only Oshie this season, leading to a career-high 22 goals. He also finished the season with a 22.9% conversion rate in all situations. The speedster has a very high career shooting percentage of 18.3% over a 281 game career, but I doubt he’s one of the best snipers in the league. Sprigings’ model agrees with me, as it has him at 12.97 expected 5v5 goals while his actual observed even strength goal total was 20.
- Viktor Arvidsson (NSH) – I’m only including Arvidsson because I found it very surprising that his expected goal total was so much lower than his observed goal total. I’m a big fan of Arvidsson and I think a lot of people had him pegged for a breakout year (Dimitri Filipovic of the PDOcast was the original leader of his fan club) due to the sheer amount of shots he produced. His breakout year came this year, as he scored 31 goals and added 30 assists while sporting a not-astronomical shooting percentage of 12.6% at all situations and 9.45% at 5v5. This is why I was surprised to see Sprigings’ expected goals model have Arvidsson at only 13.75 expected goals at even strength, opposed to his observed 5v5 goal total of 20. Maybe Arvidsson’s sample size is just too small for the algorithm to be convinced he’s a very good shooter, or maybe he did get some luck this season. Maybe it’s a bit of both.
These are some of my favourite candidates to break out or flame out in the playoffs, but there are many more highly probable candidates out there.
Regression can be either a terrible or wonderful thing, depending on who you are and which way you’re bound to regress.
*stats via stats.hockeyanalysis.com, nhl.com and @DTMAboutHeart’s expected goals data