Free Agency and Aging

Another great article by Dave Studeman on THT this morning related to aging patterns of players over time. He uses a technique where he multiplies each player's Win Shares by age and then sums them up before dividing by the sum of the Win Shares.

The interesting question he then tackles is why the avg Win Shares by age increased drastically between 1977 and 1982. Not surprisingly (once you read it anyway) is that free agency had the effect of giving older players more playing time as GM's and owners foolishly overspent on players who were past their prime.

He has a nice list of free agent contracts signed by that first class of free agents.

Players            Age       Years        Total Value
Reggie Jackson      31         5          $3,000,000
Joe Rudi            30         5           2,090,000
Don Gullett         26         6           2,000,000
Gene Tenace         30         5           1,815,000
Bobby Grich         28         5           1,750,000
Rollie Fingers      30         6           1,600,000
Dave Cash           29         5           1,500,000
Sal Bando           33         5           1,400,000
Gary Matthews       26         5           1,200,000
Don Baylor          28         6           1,020,000
Bill Campbell       28         5           1,000,000
Wayne Garland       26        10           1,000,000
Campy Campaneris    35         5             950,000

What this illustrates is how unaware the industry was at the time about aging patterns and normal career trajectories. Many of these players were past their prime and all of them were signed to contracts that guaranteed that their performance would be worse at the end than at the beginning.

Since that time sabermetrics has contributed by analyzing the subject by bringing to light the now common wisdom that peak ages are around 27 or 28 and that decline can be rapid. I looked at this question briefly awhile back and produced the following graph of normalized OPS for good players (those who racked up more than 6,000 plate appearances in their careers).



As you can see the decline is noticeable even from ages 29-31. This graph also understates the decline since the points on the right-hand part of the graph represent fewer players. In other words there is a selection bias taking place as players with lesser skills retire earlier.