As we push on into the second half of the season, a number of interesting story lines are developing from an analytics standpoint. In no particular order, here are some of the hot button topics likely to pop up as we approach the trade deadline and then the playoffs.
Toronto Maple Leafs
We’ve all heard the gory details of the decline and demise of the Randy Carlyle era in Toronto. Ugly defensive results getting worse over time eventually leading to a shift behind the bench, featuring new (old) voice Peter Horachek. A longstanding assistant in Nashville under Barry Trotz, Horachek can be found preaching the 5-5-5 philosophy whenever a microphone or camera is nearby. His articulate and detailed description of how the Leafs are looking to play as an organized five-man unit in all three zones seems like a complete change of pace from endless discussions about “compete” and “effort” under Carlyle.
Questions have already begun about how much offence the Leafs will have to sacrifice in exchange for a new-found dedication to the defensive side of the puck. So far—in a very small sample—indications actually suggest that they are generating more scoring chances than they did under Carlyle, all while drastically reducing shots against.
Unfortunately, this comes against the backdrop of a likely shooting regression. In the second half, the Leafs’ 9.6 percent 5v5 shooting percentage will drop back towards NHL average of 7.8 percent. Debate will inevitably continue if the scoring drops off though.
Montreal Canadiens
If you were wondering which team in the NHL this year is playing the percentages in a manner most similar to the Colorado Avalanche of last season, Montreal is the answer. The Habs have the 2nd highest PDO in the NHL halfway through the year at 102.0, largely floated by Carey Price’s stellar .935 save percentage at 5v5.
Price is doing his best imitation of Semyon Varlamov, who posted a .935 last year at 5v5 en route to pushing the Avs to the playoffs. But Varlamov’s save percentage and the Avs’ record regressed hard this year, and Habs fans should pay close attention to those underlying possession stats that have them sitting 21st in the NHL in Score Adjusted CF% at 49.2%—and trending down towards. The similarities between the Canadiens and their hated rivals from Toronto are scary.
Minnesota Wild
The Wild began this year with some eye-popping possession stats, keeping pace with the likes of Chicago in Score Adjusted CF%. According to War-on-Ice, they also were allowing the fewest Scoring Chances Against in the NHL for much of the first half of the year. Unfortunately, they couldn’t stop a puck to save their season (or coach Mike Yeo’s job… which he still has as of this writing). Darcy Kuemper and Niklas Backstrom don’t appear to be capable NHL goaltenders at this point, so cue mid-season trade to add goaltending help—in the form of former Coyotes and Oilers keeper Devan Dubnyk.
Everyone probably remembers Dubnyk’s epic collapse between the pipes for the Oilers last season, which predated the collapses this year of Viktor Fasth and Ben Scrivens (who haven’t actually collapsed entirely, but I’ll discuss that in a minute). Despite his atrocious stretch last season with Edmonton, Dubnyk actually has save percentage numbers that closely compare to the likes of Pekka Rinne, Antti Niemi, James Reimer and Mike Smith since the start of the 2012-13 campaign. This seems like a solid low-risk, high-reward gamble that Dubnyk is ready to reclaim the title of NHL starter, and personally I wouldn’t be surprised if it works out in their favour. He certainly didn’t look bad Thursday night.
Unfortunately those early-season possession numbers have slipped significantly so this might be a case of too little too late to save the Wild’s season, but better late than never.
Edmonton Oilers
The abysmal goaltending ride started on schedule for the Oilers again this season. In place of last season’s horse on the merry go round (Devan Dubnyk) they have been alternating between starter Ben Scrivens and backup Viktor Fasth.
Possession wise, the Oilers are up to 48.8% in Score Adjusted CF% after being 47.3% or worse in each of the past four seasons. Yet despite all the underlying improvement discussed in analytics circles, the Oilers couldn’t buy a save early on this year, and it has shown in their record. But their team 5v5 SV% is unlikely to end the year at .902, so expect some improvement in this regard.
Boston Bruins
Any team that trades away it’s second-best D-man (Johnny Boychuk) before the season begins, sees it’s Vezina-winning goalie (Tuukka Rask) post a .913 SV% through 35 starts (NHL average for all goalies—including backups—is typically around .912), misses its Norris-winning captain (Zdeno Chara) for 18 games and loses its top scorer over the past four years (David Krejci) for 19 contests should be happy to be in the mix for a playoff spot.
Now that they are approaching a fully healthy roster, the Bruins look poised to go on a second-half tear. It should also be noted that Boston has been getting exceptional performances from some of its younger players in Dougie Hamilton (21) and David Pastrnak (18) who look like solid contributors for the future.
Regression Watch
Watch for shooter who’ve posted excellent underlying stats but seen very little in the way of rewards so this season. Nathan MacKinnon is a great example. He only has eight goals on 136 shots, but sits eighth in the NHL in individual scoring chance opportunities at 5v5 (111). That puts him ahead of Tyler Seguin (110), Claude Giroux (106) and Patrick Kane (100). But he is only shooting 4.5 percent at 5v5. Last year, across both the regular season and playoffs, he shot 8.9% at 5v5. Expect him to rebound production wise.
The opposite is true for the players who have scored at levels far above their historical norms in the first half. This explains why Mike Santorelli only has three points in his past 13 games after producing 14 points in the previous 12 contests.
As the season wears on, things have a tendency to even out in the longer term. That being said, coaching changes, trades and injuries can always throw a wrench in the works. No matter how much we try to predict future performance it’s an inexact science. But, hey, that’s what keeps things interesting. Let’s enjoy the rest of the year.