Do the Brightest Stars Dim Slower?

It is well documented that baseball players tend to peak at around 26 or 27 years of age. Last year, I was talking a little baseball with a person who knows the game well. At some point in the conversation, the aging process of position players became central to the discussion. I suggested that the Padres should be wary of signing or trading for any star player already in his 30s-even early 30s, because an expected decrease in skill may leave the club paying for past performance rather than future performance. The person I was talking to suggested that star players actually experience a much longer peak than the average ball player. Therefore, acquiring a star player in his early 30s could actually be a good bet for the Padres, as the star will be likely to maintain his production, or at the very least experience an extremely gradual decline.

At the time, I did not pursue the argument any further, but continued to silently contemplate the concept that the best players may age better than the average player. I have no doubt that hall of famers continue to be extremely productive well into their 30s. There is no way that they would be able to accumulate the longevity statistics necessary to be voted into the hall without sustained excellence. At the same time, there is no reason to believe that just because they played great baseball at 33, they were not even better at 27.

The biggest problem I had with the argument was that it is very difficult to project the players that will age well, and who will completely fall off the map. Sure players with greater athleticism have been shown to age better than huge, clumsy first baseman/DH types such as Cecil Fielder, but it is very difficult to make predictions for an individual player. Most of the players fans remember as stars were able to sustain their excellence, but for every Paul Molitor who seemed to grow better with age, there is a Ruben Sierra who looked like a star early in his career, but struggled after experiencing early success.

Since the conversation began with the question of whether the Padres can safely sign or trade for “stars,” in their 30s, I decided to construct a test that examines the careers of players who at any given point may have been considered a “star.”

The first step was to define the parameters. I chose to define a “star” as a person selected to at least 2 allstar games-at least one of which occurred in the 90s. I examined stars from the 90s because these players were able to take advantage of relatively modern training and medical technologies. In addition, most of these players are now retired or at the very end of their careers, which allows for a longitudinal view. I included only players elected to at least 2 allstar games because this removes fluky players who may have only had one good half season, or were only elected because their team needed a representative (sorry guys, but Jon Jaha was removed from the data set).

Skill was defined by WARP (Wins Against Replacement Player). WARP has the advantage of measuring both offensive and defensive production. About 1/3 of the way through my data collection, I realized I probably should have used WARP 3, but I did not really feel like starting over at that point, and I am certain that the results were not affected by the subtle differences between WARP 1 and WARP 3.

I recorded the age of each player when he achieved his career high WARP, the lower and upper bounds of each player’s best three year WARP, and each player’s cumulative WARP from their age 26-28, 29-31, and 32-34 seasons. I calculated the mean, median, mode, and standard deviations for each category where applicable. I present the results in the table below. In addition, feel free to download the excel spreadsheet that contains the raw data and calculations.

Results (I can't seem to get the formatting right, so please scroll down):

	Age: Best WARP	Age: Best 3 yr. WARP Low	Age: Best 3 yr. WARP High	WARP 26-28	WARP 29-31	WARP 32-34
Mean	27.867	26.965	28.965	18.366	16.073	11.944
Median	27	26	28	18.3	16.3	11.1
Mode	27	26	28
Std. Dev.	2.987	2.982	2.982

It is pretty clear that even the stars experience their best play between 26 and 28. From 29-31, there is a gradual decline (12.48%), and then an even more dramatic decline from 32-34 (an additional 25.69% or 34.97% from the age 26-28 seasons). These results are extremely consistent with the numbers seen in the general baseball population.

My conclusion is that although some players maintain an extremely high level of play well into their 30s, that in general, even the stars demonstrate severe slippage from their peak level of play, and that slippage tends to begin between ages 29 and 31. This fact suggests that making a large investment in a 30 year old star is a very risky proposition. Although it is likely that the star will continue to be a serviceable regular worthy of consistent playing time, the team cannot expect the player to duplicate the production from earlier in his career, and should plan accordingly.