I have a friend in the industry who swears by this approach. It depends on how hyper-focused you're. Some of these guys go to games intent on focusing on just a couple guys. They watch them almost exclusively, so yes I could sit at a Frolunda game and barely notice another prospect unless he is having a spectacular game that makes you notice. By Hakan's stories on Datsyuk if I remember right that is how he got noticed, he was there to see somebody else and actually just started watching Datsyuk. That will happen and my friend says in that case it happened for a good reason, but he actually goes into a game with a clear objective on what to watch and when to watch and it isn't the same way most of us are watching a hockey game.
He thinks the sweet spot is 3 to 5 games. Flip on some film and talk to coaches, scouts and whatnot, make sure you didn't witness the four best games of the kids life. But he found early in his career you weren't paying enough attention to enough players and you started nitpicking guys when you focused too much on them. There is a balancing act in his opinion. He joked that I must have saw all 5 bad Rossi games this year when we talked last and to dig deeper...
Absolutely great to hear this. Been doing the same, and know that best scouts do the same.
That’s fine, but when building statistical models on kids 8 games is totally irrelevant imo.
Why build a statistical model based on 8 games you watch?
You go to the game for the EYE-TEST.
And build a statistical model based on his whole season, and look if it will fit on your eye-test or not.
I've achieved tremendous results by working like this (in a keeper league).
EDIT:
Nowadays there's so much of shooting-based data available, that you can take that from stats straight on. But that's definitely not the whole thing.
If you like to track some data also by eye-test, is the passing plays, who can't be tracked from stats anyhow, until the Smart Puck -solutions and data come available in prospect league. Track them and merge to shooting data. then you have almost everything. But defensive data is still a different world....
Also intangibles are another thing, like, you can't see a good SCREEN play from any data, because the player doesn't always touch the puck. If tip-in yes, but sometimes the play (blocking goalie's wiev) is the most important for a goal to happen, when players doesn't touch anywhere. Any data doesn't have anything about those kind of plays anywhere. Again, until the Smart Puck -comes... Player locations (with location sensors) matched to shot timing (with stick sensors), then maybe....
Then the whole scouting world will change. It's very near. In Finland we have already that system in use and developing, but certain (most important) stats are kept secret from public... wonder why. It's super lucrative solution.