I was asked this question on Hey, Bill.
The Dodgers have just signed Matt Kemp to an 8-year deal at $20 million per. What are the chances that an All-Star caliber player at the age of 27 will be an All-Star caliber player eight years out?
Asked by: bobfiore
Answered: 11/16/2011
I wrote an answer, but the stupid computer refuses to allow me to copy my answer into their box, so I’ll have to answer it here. . .sorry to step on Dave’s article. Anyway. .
Kemp was actually 26 in the middle of the 2011 season, so we’ll treat him as a 26-year-old. Starting with all players who are 24 to 28 years old, who create at least 4.50 runs per 27 outs, who have season scores of at least 200, who have played 500 to 1200 games in their careers, and who have runs created per game for their careers between 5.00 and 7.50. . .that gives us 956 players, which is too many to be useful. We trim it down to 25 to 27 years old; that gives us 612 players. Kemp created 8.28 Runs/27 outs this year, so we’ll trim the list to a range of 6 to 11 there; that gives us 472 players. Kemp’s season score was 462, so we’ll increase the minimum there to 250; that gives us 370 players.
Kemp had 39 homers and 40 steals, so we’ll eliminate anybody who had less than 10 homers or less than 10 steals. That gives us 151 players. Let’s eliminate anybody who did this less than 8 years ago. . ..well, no, that would eliminate Carlos Beltran, who appears to be one of the best comps, with 38 homers and 42 steals in 2004. Let’s eliminate those from 2005 onward; that leaves us with 131 players, still a few more than we want. We’ll eliminate anybody who had less than 100 runs created; that gives us 117. We’ll eliminate the 19th century players; that leaves 115. Kemp in his career has 128 homers and 144 steals, so we’ll eliminate anybody who is under 50 in either category. That leaves 70 players. Let’s increase the Season Score requirement to 300; that leaves 57 players
The list of 57 players includes, however, Carlos Beltran in 2002, Carlos Beltran in 2003, Carlos Beltran in 2004, etc., so for those cases we’ll include only the year in which the player was 26 years old. We have two cases, Andre Dawson and Bobby Bonds, in which the player is on the list at ages 25 and 27, but not at age 26. For those cases, since Kemp is much closer to being 27 than 25, we’ll use the age-27 season.
That gives us 43 players who seem like a valid peer group for Kemp; chronologically, those players are Bob Meusel (1924), Heinie Manush (1928), Joe Gordon (1942, MVP), Duke Snider (1953), Hank Aaron (1961), Carl Yastrzemski (1967, MVP), Jimmy Wynn (1969), Roy White (1970), Bobby Bonds (1973), Reggie Jackson (1973, MVP), Mike Schmidt (1976), Dave Parker (1978, MVP), George Brett (1979), Dave Winfield (1979), Andre Dawson (1982), Dale Murphy (1982, MVP), Pedro Guerrero (1983), Rickey Henderson (1985, made Don Mattingly the MVP), Ryne Sandberg (1985), Von Hayes (1986), Lloyd Moseby (1987), Kirby Puckett (1987), Darryl Strawberry (1988), Andy Van Slyke (1988), Bobby Bonds (1991; actually WASN’T the MVP that year), Jose Canseco (1991), Ruben Sierra (1991), Roberto Alomar (1993), Gregg Jefferies (1993), Reggie Sanders (1995), Gary Sheffield (1996), Bernie Williams (1996), Raul Mondesi (1997), Chipper Jones (1998), Shawn Green (1999), Ivan Rodriguez (1999, MVP), Johnny Damon (2000), Darin Erstad (2000), Derek Jeter (2000), Magglio Ordonez (2001), Scott Rolen (2001), Carlos Beltran (2003) and Alfonso Soriano (2003)
Let’s use a Season Score of 200 as the standard of a season that you wouldn’t feel totally ripped off if you paid $20 million for it. …"200" in 2011 was Adam Jones of Baltimore--.280 with 25 homers, 12 steals, 83 RBI. "190" in 2011 was B. J. Upton (.243 with 23 homers, 81 RBI) or Carlos Quentin (.254 with 24 and 77). If you play at the level of Adam Jones (in the studies that follow) we count that as a "successful" season; if you’re at the level of Upton or Quentin, it’s not a successful season.
Of the 43 comparable players, 32 met the standard of a successful season in the next season, but actually we should count it as 34 because Jefferies and Alomar would have met the standard if the 1994 season had not been terminated by a strike. . .so let’s say it is 34/43 (or 80%) in the first year of the comparable period.
28 of the 43 players had successful seasons in the second year going forward, but we need to count it as 31 of 43 because Winfield and Brett were prevented from meeting the standard by the 1981 strike, and Joe Gordon by World War II, so that’s 72%.
30 of the 43 met the standard in the third year, which we’ll count as 32 of 43 because of Sierra (1994 strike season) and Joe Gordon (1945). That’s 74%.
29 of 43 met the standard in the fourth year, which we’ll count at 30 of 43, so that’s 70%. . .first four years are 80%, 72%, 74%, 70%.
26 of the 43 met the standard in the fifth season (60%); even the superstars are starting to flake off here. George Brett did not meet the standard in 1984, Duke Snider did not in 1958, Yastrzemski did not in 1972.
27 met the standard in the sixth season (63%). Only 18 met the standard in the seventh season (42%), and only 18 met the standard in the eighth season, which is the season you actually asked about. So. …80, .72, .74. .70, .60, .63, .42, .42; the Dodgers can expect Kemp to have 5.05 successful seasons over the eight-year contract. Let’s call it five.
Extending it on out, 14 of the players had successful seasons in the 9th following season; however, that’s 14 of 41 since we haven’t yet reached the 9th season for Beltran or Soriano, so that’s 34%. Only 6 of the 41 players had successful seasons in the 10th following season, although others were close and some had successful seasons AFTER the ten-year window, but not in the tenth season, but anyway, that’s 15%.