[JeffR]

Researcher: JeffR

So, time for a somewhat lengthy explanation on where I am at the moment with the historical db:

Anyone who debuted between 1947-48 and 1979-80 is in the game right now. The ratings are pretty rudimentary at the moment, but they do produce reasonably good results. They'll get more precise as the detailed player records are filled out in the database, but that isn't going to happen until every player has the basics done. And I'm the bottleneck on getting that done: having done the first 40+ years myself, I want to carry it through so I do the more recent players, too. That's mainly out of a desire to keep the basic framework of the ratings consistent across the entire database - I get inconsistent enough on my own, it would get really messy with multiple pairs of hands at work. (And, to a lesser extent, it's because I have fun working on it - after spending hours writing documentation, testing, dealing with email and PM's, and moving numbers around in the main database, studying the career of Morris Mott is nice stress relief.)

Once I've got every single NHL player from 1947 forward rated in a basic way, we're going to split the database up and start doing the detail work that gives the players more personality - making Bobby Clarke great at faceoffs, Bob Probert a terrific fighter, Harold Snepsts slow enough to time with an hourglass, and so on. That's when we should start to see really impressive results. I've got several volunteers already from the current research team who want to work on those, so I think we're probably covered for research help at the moment, but I'll post if we need more people to help with that.

With things at the stage they are, I can't really use detailed information on the players just yet. If I've got something wrong in general terms - you remember a player was good defensively and I've got him modeled as a goal-scoring floater - by all means let me know and I'll fix it. I just can't do anything (yet) with the knowledge that he wasn't a great fighter and couldn't throw a check to save his life.

One specific thing that I particularly welcome: better data on what NHL teams owned a player during seasons he didn't actually play any NHL games. If you start up a game as the 1974 Bruins and see a guy there that you know was sold to Atlanta that offseason, I want to hear about it. The bulk of the problems will be with Original-6 era guys, where the transaction data isn't available for a lot of the times they moved unless you dig through vintage THN's or local newspapers. Case in point, a guy who had no NHL transaction history I could find prior to 1967 but had a long minor league career up to that point. As it happens, his grandson is one of our researchers, and after talking with him confirmed that he'd actually been Hawk property in the early 50's. So now, with the "historical debut" option turned on, he shows up in the game then, instead of 15 years later. That's the kind of information that's very valuable to me.