Bear of Bad News
Your Third or Fourth Favorite HFBoards Admin
- Sep 27, 2005
- 13,842
- 33,940
Counter point:
Loving and frequently using analytics is mainly so that people who never played and don't watch a lot of hockey can go around saying "I know what I'm talking about, this player is better than that player, this team is better than that team".
In my experience, talking about teams/players from even the late 2000's-mid 2010's it's easy to see who actually watched these guys play and who is just checking out the stat-line. The stat-line has gotten more detailed but it's still the same concept. Analytics give you an idea of what happened but nothing replaces watching the games. I posted this opinion a few pages back but I'd bet over 90% of us on HF aren't watching enough hockey to have substantial opinions on the players/teams across the league. If you're not watching you can't fully know and hockey is too contextualized to pull substantial conclusions from the data.
I find anything/60 to be next to useless and many goalie stats to still not be as useful as a simple SV% and GAA metric. I'd go as far as to say the "standard" stats on NHL.com are more than enough.
I can't speak for everyone, but I can tell you that my friends and colleagues who do this work professionally got into this because they watched a lot of hockey and wanted to apply their knowledge to better understanding. The people I've worked with typically have games on from 7pm ET until 1am ET and then watch the highlight shows on repeat while they're working. Caveat that I haven't been involved in hockey analytics professionally since the pandemic began because I had a second kid (and now have three) and my skills have eroded to the point where I'm not useful, but people do not choose these low-paying jobs unless they have a strong interest in the sport they'll be working in. About a decade ago, a recruiter reached out to me with an offer from a major professional team to be half of the analytics team. The pay being offered was about what we were paying an entry-level actuarial student with one or two exams passed.
One of the biggest problems with sports analytics these days is that a lot of the key voices are. Not. Good. Communicators. They don't know how to frame a value proposition, they don't know how to view material from an audience's perspective, and they don't understand the point of concise persuasive writing. I still read Bill James books multiple times per year, not because of the math, but because he reminds me just how much entertaining writing pulls in an audience and makes the point more understandable.
The other big perception problem with sports analytics is that the stuff in the public domain is of two types: (1) people trying to get hired by publishing what they can do with public domain data, and (2) things teams allow to enter the public domain because they've developed better proprietary things. When you say that sports analytics have no value, remember that the professional teams agree with you - and that's why it's in the public domain. I'm exaggerating for effect - some of it has some use - but the best stuff is locked down until it's no longer the best stuff.
Anyhow, this is more than you wanted to read but you've already read it so now you're stuck, I guess.