Back at the start of the season, Tom Timmermann from the St. Louis Post Dispatch wrote about "moneypuck" and the rise of analytics in the NHL. More specifically, he wrote about how the St. Louis Blues and Hitchcock have been using analytics. As the season progressed the Blues had some very successful line combinations, like the STL line for instance. But when the team would struggle within a game or during a slump Hitchcock would experiment with new line combos. Some clicked but most of them just seemed odd. As these different line combinations kept happening the opening two paragraphs from Timmermann's article kept bouncing around in my head.
At the end of each period, Blues coach Ken Hitchcock walks back to his office. The walk doesn’t take much more than a minute, but by the time Hitchcock gets there, he’s heard the ding on his phone and he knows there’s a message waiting for him.
It’s the Blues’ analytics breakdown for the period, a rundown of what worked and what didn’t, what matchups to promote and which ones to avoid. (The Blues are purposely vague about what’s actually in there and who is sending it.) And that information is something that Hitchcock takes very seriously.
So Hitchcock is using analytics. Specifically he is using them to determine on-ice matchups and these match-up reports are something that he is fully committed to. Of course we don't know what sort of information these match-up reports are providing nor do we know the company that is providing them. But if Timmermann is being accurate in describing Hitchcock's commitment to them, this obviously goes a long way to explaining some of the line combos Hitch has been putting on the ice throughout the course of the season and into the playoffs.
I scraped the TOI data for all 12 forwards from the first three playoff games for the Blues. I visualized it in the dataviz below so you can play around with it yourself to see how Hitchcock has been rolling his lines during the playoffs. I know the visual is a bit clunky. It usually helps to filter a few players at a time. You can also click on a player's name to highlight them and search vertically for other forwards on the ice at the same time.
The chart starts off with Backes, Berglund, and Jaskin because it looks like that was a pretty consistent line combination Hitchcock used during game one this series. Then I went over to the new Puckalytics site and used its SuperWOWY (with or without you) page to look up Backes, Berglund, and Jaskin. During the course of the season, they played roughly 60 minutes together. During the times they played together this trio had only a 42% Corsi For. However, they did have a 1066 PDO with a 10% on-ice shooting percentage. Lucky much?
So this really begs the question. What data is Hitchcock being provided in his match-up reports? Because on the surface it might seem like the Backes-Berglund-Jaskin line is a good line combo, but in reality they were nothing but lucky with worse possession numbers together than they had when they were apart.
Now this is just scratching the surface and just a quick look at the first three games of the playoffs. I will be working on a more substantive look at this issue during July after the Blues finish playing hockey.