Thursday, January 6, 2011

A new projection system, I think?

A week or so ago, I took a look at some of my projections from last season. Some were close, others weren't close at all. So I decided to take a look at my formulas. My hitter projections last year were completely numbers based on what I learned players typically do from year to year. Many of these players are not typical though. I have to do more.
What I ended up doing is amending what I had by adjusting for each players age to see how much improvement he should have compared to his league. I created a new set of adjustments. I averaged the formulas I use for each league to create a formula for each level. This isn't the most accurate thing, but I'm essentially trying to create a crystal ball so 100% accuracy is unattainable and pointless to even attempt.
Over the last few days, I have spent dozens of hours trying to make this work, and honestly, I think my brain is a litte fried, but I think I have something.
What I have is essentially a stat line of what a player can be reasonably expected to do in the majors. I still have to look more deeply at the results, as I just finished this morning at about 3:30 AM. I had a busy day, so I haven't looked at it yet. I will probably take a look here yet tonight and make some of the info available tomorrow or the next day, if it makes sense. I'm posting this now, before analysis, because I am excited about it and quite proud as well.
I hope it's as interesting as I think it is.