I don't -- I figure the folks at Baseball Prospectus have a slightly more powerful database and program than I do, so I let PECOTA have all the fun for me.
[Edit: Probably the most effective way is to figure out a set of career paths and apply the models to individual players, projecting future performance based on similar player profiles from the past.
Or you could go with the classic 50% weight on last year, 33% on two years ago, 17% on the previous year, but that doesn't really account for declining performance.]
[Second edit: Peak year for batters is typically 26. Baseball Prospectus had a nice article on this a while ago. If you're a subscriber, you can find it here:
http://baseballprospectus.com/article.p ... cleid=4464
To summarize, the peak age is as follows for these positions:
C - 27
1B - 26
2B - 25-26
3B - 26-27
SS - 26
LF/RF - 26-27
CF - 25
And the aging models differ for positions. First base follows an average career path, corner outfielders drop off sooner, CF and middle infield peak sooner and higher but drop off more, catchers rise and fall very quickly with that peak the only time they touch average, third base stays better a bit longer then drops off at age 32 or so.]
Your wisemen don't know how it feels to be thick as a brick...