Erboes wrote:My motivation for this is not to prove I'm right, just to find a more useful method of drafting. Here are my two findings:
1) A player who has shown substantial improvement one season over his previous is unlikely to better those numbers the next season or even to repeat them.
2) A player who has a substantial drop in numbers one season compared to his previous season is likely to improve on his numbers the next season over the one previous.
How can you dispute this?
I'm not sure anyone is disputing these broad generalizations, only cautioning that they are simply generalizations. Yes, you are right. I agree. But I'm not going to start drafting guys who had drop-offs over those who improved the previous season. There are far too many other factors to consider than to use this as a drafting strategy.
Consider a player in category 1: Hank Blalock. I'm very excited about him for next season. He posted a substantial improvement in 2003. Should I expect a step backwards now? Probably not - because he's so young and seems to be in a developmental upswing.
Or consider a player in category 2: Ray Durham. Big dropoff in production. Should we expect him to post better numbers? Let's say probably. Does this make him a valuable 2nd basemen, or worth taking in the first 5 rounds? No.
Again, I'm not disagreeing with the generalizations. I think it is a very wise strategy to look at players who had drop-offs and to consider WHY they had drop-offs. This is a great way to find gems in the draft. Burrell and F. Garcia are two players that have a shot at performing better in 2004 than in 2003.
Would I take Berkman over Sheffield? I might. I wouldn't take either in the 1st round, and wouldn't let either slip to the 4th.
But look, you've retreated from relying on your statistical data to relying on fantasy baseball vernacular. I wish I had the time to run the numbers, because then I'd try and do it. I think it would be very interesting to do a more rigorous test of these generalizations. Then we could look for trends by age, by position, by injury history, etc. I'm not trying to criticize what you were doing, I was only pointing out that to do it right would require a more rigorous test.
"The game has a cleanness. If you do a good job, the numbers say so. You don't have to ask anyone or play politics. You don't have to wait for the reviews." - Sandy Koufax