Tuesday, May 14, 2013

WAR by Age--An Analysis

This chart has flummoxed me for years:






















There are two axes so that the OPS line doesn't overwhelm the other three values. Every time I went through the data I came up short when I saw these graphs--there's no dramatic drop-off over time. The prior post described why this is--as players age and performance ebbs, their numbers don't decrease as much as they no longer HAVE numbers because they're out of baseball. 

This chart is a much more accurate representation of aging and diminished performace:
  





















Another two-axis chart, the number of players by age from 1901-2012 on the left and average WAR by age for the same time span on the right. For example, from 1901-2012, there were just over 500 players who played at age 20, with an average WAR of -.6. I'm surprised that the peak in player age (26) doesn't coincide with the peak in WAR (30). The number of seasons after age 30 begins to drop precipitously, and any little upward blips in the WAR line around age 39 are due mainly to the scientific experiment that was Barry Bonds.

The decline in ability also accelerates with age. After peaking around age 30, WAR values begin a gradual decline that grows greater around ages 35-36. The people that perform well after age 36 are simply rare breeds. NOT including Barry Bonds, since 1990 the best season by a player aged 37 or more was Edgar Martinez with a 5.6 WAR for the 2000 Mariners. This chart shows season WAR values for all players from 1990-2012 aged 37 or greater:

The vast majority of these players had a WAR of 1 or less. There are 560 seasons shown in this chart, of which 158 had a WAR greater than 1--WAR values between 2-5 are solid seasons, 5-8 All-Star type seasons and 8+ MVP-type seasons. The stellar seasons are not represented here, suggesting that even while players may have continued to play as they aged more than in the past, they weren't any more successful. Here's the breakdown by position, as well as how many had more than 300 PA in a season:

277 seasons of 300 or more plate appearances, right around half (49.5%). They're primarily in positions that don't require speed or agility, and it's pretty obvious that by the time a catcher reaches 37, they're a backup. I just ache THINKING about catching at that age, let alone doing it over the course of a season.















This last chart is the fun one--it shows the best seasons by age since 1901 using WAR. In the rare instances where it mattered, at least 100 plate appearances were required:
  


























Other than the young ages, no real surprises here, but it's fun to see it all in one place. It also shows just how amazing 2012 was to have the best seasons ever by 19 and 20-year-olds, even with small sample sizes.

This post and the prior one together tell a pretty simple story--even though players may be playing longer, they're not any more effective than they were in the past in the aggregate. Of course, individual players defy this, and that's the challenge every general manager faces, to differentiate between a breakdown and the potential for another solid season. I remember reading an essay prior to the 1990 season that discussed Keith Hernandez, who had signed as a free agent with the Indians at the age of 36. The essay made the point that players rarely have great seasons at that age, and that was prior to the wealth of information available to lay people like me. It ended up being Hernandez's last year in baseball--145 PA and batting a robust .200. One of the best players of his generation, done at 36.

The day is coming sooner rather than later when the 34 or 35-year-old baseball player will be as rare as the 30+ running back in football. I will be very interested in how many Josh Hamilton-type contracts are forthcoming in the future. I thought Prince Fielder was a big enough gamble after the 2011 season when he was 27, but that was more because I've never been convinced that he wouldn't be the next Mo Vaughn--MVP at 27, overpaid at 31, injured and washed up at 35, but so far he's performed. I just heard on the Boers and Bernstein Show on the The Score (click the link to the right to listen) the question of what Bryce Harper and Mike Trout should do--sign a below-market (but still extremely lucrative) contract like Anthony Rizzo with the expectation that they can sign a mega-deal in another 5-6 years when they'll still be around 26-27, or wait for the BIG contract? Terry Boers mentioned he'd sign NOW, lock in the money and then sign another big contract. Or, they can do like Evan Longoria apparently did and renegotiate the contract he signed during his 2008 rookie season. 

The money is going to begin to move to the younger players, and I strongly suspect the days of contracts like Adam Dunn's aren't just history, but ancient relics that people in the future will pore over and ask themselves what it means. There will always be teams that throw money around like the Yankees and Red Sox, but they're the exception and not the rule, and it's NOT because they have more money. The revenue available to most teams will only grow as teams develop their own TV networks like the Yankees' YES Network or the Dodgers network, but it won't necessarily go to 32-year-old position players. In fact, I'll go so far as to say that's where it WON'T go except in very special circumstances that will be fewer and fewer every year.

1 comment :