I'm always willing to listen to other ideas. I don't know anything about those metrics, so I've been curious.
In the end, twabby's argument is that the GAR model has a value assigned to Brown and they're comparing that metric to other players and assessing Brown like that.
Some of the things that twabby doesn't care to look at, I do. I value that stuff (like active sticks, hustling, blocking shots and positioning) because I was raised in the game and played for a very long time. The GAR stuff, from what I've read, is an estimate based on statistical models. It has grey areas in it and I wouldn't evaluate a player solely on it yet, but that's because I don't know how those numbers are assigned. I don't understand the bias for valuing that number above other things because I'm ignorant when it comes to that stat and what it really means.
It's very hard to compare player to player because there is way too many different variables in hockey. This is not baseball, advanced stats will never tell us the same thing. In hockey, you have 10 players and 2 goalies battling for a puck at the same time. Quality of teammates, quality of competition, trailing, leading, end of periods/games, start of periods/games, zone starts, roles, defensive missions, matchup missions, special teams, chemistry, injuries, coaching, etc etc etc, so many factors.
Sorry to say in that bluntly, but relying entirely on "statistical models" is stupid. You do that and you end up being Byron Bader.
The only realistic way to look at analytics is to take it as ONE information, exactly like for any other stat, this is not and will never be the "be-all and end-all"
If Brown is so good why did the Sens dump him when it looks like they are trying to win now?
Simply because their GM is stupid (this is based on 6 years observing him). I don't remember saying that for any other NHL executive in over 3 decades
Giroux is the bigger name (and even better) so him and his salary replaces Brown, all Sens fans accept that. The internal budget is obviously important in Ottawa, above anything else. Problem is Pierre Dorion doesn't seem to have any notion about "asset management". He doesn't understand what "buy low sell high" means, which is why you'll see him let a 25 y/o guy like Anthony Duclair walk as a UFA, for example.
This was the worst timing to move Brown's salary (see Pacioretty trade, obvious buyer's market) but Dorion still did when there was NO RUSH to do it. I doubt the entirety of Brown's 4 M$ salary this season had to be paid in full shortly.
#moneyinmoneyoutFTW!
Totally agree, though he should still know better. We had Jay Beagle here for years -- another actual defensive standout that analytics weren't kind to because of his deployments. He was our only decent faceoff guy, so he got a lot of extra defensive-zone and shorthanded assignments just to shelter other centers on important draws. Nowhere near the offensive upside of Brown, but nearly identical two-way and special teams play, overall motors, and the kind of character effort that make their teammates and fans adore them.
But look at fancy stats with the wrong tilt of your head and they look unremarkable in spots.
Fancy stats are useful tools for evaluating talent, but there is no model that tells the whole story, and this is an example of how it can go wrong. Connor Brown is a very good hockey player and worth every cent of his contract.
He's an instant fan favorite. This is what he did in his first game in Ottawa
Not that it's spectacular or anything but a very nice visit card of things to come. A puck hound looking to pounce on every little mistake from the opponent
Advanced stats don't tell the whole story. Patrick Wiercioch used to be an analytics darling. 'nuff said
