The Trade Value Series Skews Young (Again)

Today, we wrapped the 2019 Trade Value Series. The series always offers a number of a interesting insights into the industry’s thinking leading up to the deadline, and serves as a reminder that younger players tend to have considerably more trade value than their older peers. There are multiple reasons for that phenomenon. First, younger players are cheap. Owners and players have agreed to a system that pays players around half a million dollars for the first three seasons of their major league careers, followed by another three or four years of arbitration during which salaries increase gradually, but are only guaranteed for a single season at a time, limiting risk for teams. Then, after six full seasons in the big leagues, players hit the free agent market, where every team is welcome to bid for a player’s services. As a result, players who reach free agency tend to have much higher salaries than their younger teammates. It stands to reason when determining trade value, then, that, assuming an equal level of play from a younger and an older player, teams would value the younger player more highly because said player is cheaper.

The logic above can be seen pretty clearly in this year’s Trade Value Series, as well as those of the past decade. The graph below shows the average age of the players featured in the Trade Value Series over the last 10 years, with the first eight installments of the exercise conducted by Dave Cameron and the last two performed by Kiley McDaniel.

The players in this series have always been on the younger side, with this year’s average age of 24.8 years old right in line with the 10-year average of 25.2 years old. The phenomenon discussed in the opening paragraph generally explains the Trade Value Series, and illuminates the reasoning behind teams further emphasizing their farm systems, player development, and the unfortunate practice of service time manipulation. But the current collective bargaining agreement doesn’t completely explain why younger players are more valuable than their older peers. While the salary structures and economic incentives created by the CBA haven’t meaningfully changed in the last 10 years, the age-based demographics of the game have shifted drastically. Why? In part, because younger players are a lot better than they used to be.

The graph below shows a rolling five-year average of the share of WAR by age dating back to 1984. The groups are split into three categories: 24 and younger, 25 to 30, and 31 and older:

Position players considered to be in their prime represent the largest portion of WAR over the years, generally moving between 55% and 60%, and seeing an uptick over the last few seasons. Players in their 30s have seen much wider movement, with a fall throughout the 80s and a rise in the 90s followed by the downturn we’ve seen over the last decade. There are a couple potential explanations. The first is expansion. As MLB moved away from its last expansion in 1977, older players’ share of WAR fell. It began to rise with the two rounds of expansion that occurred in the 90s, and has fallen since the league has remained at 30 teams for more than 20 seasons. While certainly speculative, the fall of production as it relates to older players also coincided with the implementation of PED testing.

While the share of WAR for older players has gone down, younger players have picked up the slack. The changes are even more stark when looking at the quality of play. The next graphs shows WAR per 600 plate appearances for the same age groups:

This change hasn’t come about because teams are being more careful or selective when it comes to players 24 years of age and younger, either. The younger group has performed better as it has received an increasing share of plate appearances while older players have gotten worse as they have lost playing time:

As teams have gotten better at scouting and developing players, and as an increasing number of players have come from international markets, the available talent in baseball has increased, which has had a hugely negative effect on older players. This effect has been seen much more among position players than pitchers. The graph below shows the share of WAR by age for pitchers.

We’ve seen significant medical advances for pitchers, and teams have attempted to take greater care for arms, which could allow pitchers to age better than they had previously. As starters throw fewer innings, it could serve to reduce strain on arms, as well as affording older pitchers who are conditioned to pitch deeper a slight innings advantage. Of the 15 individual pitchers’ seasons exceeding 210 innings the last three years, 10 have come from pitchers at least 30 years old or older. Increasing the number of relief innings also provides an avenue for aging pitchers to remain in the league and potentially at a high level.

As for the position players, to explain how the older players have been moved out of the league, let’s consider a made-up baseball league with about 600 players who have true talent levels ranging between 0 and 8 WAR. For the sake of this exercise, we’ll take a very aggressive stance on aging and assume that players gain one WAR per season up until 25, maintain that level through their age-30 season, and decline by one WAR per year thereafter. Here’s how the age and talent distribution in that made-up league might work to start with:

Roster Spot Scenario By Age and WAR
AGE 8 WAR 7 WAR 6 WAR 5 WAR 4 WAR 3 WAR 2 WAR 1 WAR 0 WAR Roster %
22 1 2 4 6 7 8 4.5%
23 1 2 4 6 7 8 9 6.0%
24 1 2 4 6 7 8 9 10 7.6%
25 1 2 4 6 7 8 9 10 11 9.4%
26 1 2 4 6 7 8 9 10 11 9.4%
27 1 2 4 6 7 8 9 10 11 9.4%
28 1 2 4 6 7 8 9 10 11 9.4%
29 1 2 4 6 7 8 9 10 11 9.4%
30 1 2 4 6 7 8 9 10 11 9.4%
31 1 2 4 6 7 8 9 10 7.6%
32 1 2 4 6 7 8 9 6.0%
33 1 2 4 6 7 8 4.5%
34 1 2 4 6 7 3.2%
35 1 2 4 6 2.1%
36 1 2 4 1.1%
37 1 2 0.5%
38 1 0.2%

If every season welcomed a new crop of 22-year-olds with the same talent as the year before, the age and talent composition of the league would remain almost exactly the same in perpetuity, with some potential variation in the ages of the replacement players. However, even small changes in the pool of 22-year-olds could have a huge effect on the older population of players over time. If we added a five-win, four-win, and three-win player to the group of 22-year-olds entering the league ever year, adding about 10% to their total number of players and 30% to their WAR totals, those three players would take three replacement-level jobs, presumably from the oldest among the group. Repeat the process year after year and it looks like this:

After 10 years, the final graph looks like this:

After 10 Years: Roster Spot Scenario By Age and WAR
AGE 8 WAR 7 WAR 6 WAR 5 WAR 4 WAR 3 WAR 2 WAR 1 WAR 0 WAR Roster %
22 2 3 5 6 7 8 5.0%
23 2 3 5 6 7 8 9 6.5%
24 2 3 5 6 7 8 9 10 8.1%
25 2 3 5 6 7 8 9 10 11 9.9%
26 2 3 5 6 7 8 9 10 11 9.9%
27 2 3 5 6 7 8 9 10 11 9.9%
28 2 3 5 6 7 8 9 10 11 9.9%
29 2 3 5 6 7 8 9 10 11 9.9%
30 2 3 5 6 7 8 9 10 11 9.9%
31 2 3 5 6 7 8 9 10 8.1%
32 1 2 4 6 7 8 7 5.7%
33 1 2 4 6 7 0 3.2%
34 1 2 4 6 0 2.1%
35 1 2 4 0 1.1%
36 1 2 0 0.5%
37 1 0 0.2%
38 0.0%

After a decade, a small change in the flow talent costs players who are 31 years of age and older about 3% of their roster spots. While they will gain some back as the new young talent ages, they won’t gain them all back, and if younger players continue to get better instead of staying at the same rate, any of those minor gains will be eliminated. With a finite number of roster spots, it’s easy to see how expansion combats the growing number of better, younger players to help veterans keep their jobs. Young players might gain some of those extra roster spots in an expansion setting, but 50 more roster spots would help the veterans a lot more by lowering the overall level of play to help halt declines and ensure jobs for those closer to losing their spot in the game.

As the age of players and the quality of their contributions change, the pay scale remains static, with veterans of at least six years of service time taking home two-thirds of the salaries despite providing under 30% of contributions:

WAR and Salary By Service Time in 2019
Service Time % WAR % Salary
0-3 42.0% 5.5%
3-6 30.2% 27.2%
6+ 27.8% 67.3%

While it is important to recognize the union agreed to this financial structure, it is equally important to understand that to the extent it worked in the past, the pay structure in the last few CBAs remaining largely unchanged has exacerbated the gap between player pay and their play on the field. The Trade Value Series is almost always going to favor young players, and it is going to favor the best players, but changing age demographics in the game makes it increasingly difficult for veterans to crack the list.

Craig Edwards can be found on twitter @craigjedwards.

Newest Most Voted
Inline Feedbacks
View all comments
4 years ago

That last graph makes a strong argument for the concept of arbitration in baseball.

I’ve spent a lot of time (way more than I should) trying to figure out what should be tweaked in baseball’s economics to balance money flows better, but there are so many competing concerns tied up in it, it’s hard to “solve” one issue with breaking others.