Hamilton’s comparables, age 26-31 and 32-36

Josh Hamilton‘s new contract covers age 32-36. So I set out to find players from the live-ball era who were comparable to Hamilton from age 26-31, and see how they had performed at age 32-36.

Before proceeding, there’s one point that I failed to state clearly in the Greinke post: I am not analyzing the financial side of Hamilton’s contract. It’s too soon to gauge the impact of the recent revenue surge on the average cost per WAR on the free market, and there are too many other value factors specific to the situation. I’m just looking at on-field performance.


Since the vast bulk of Hamilton’s value comes from swinging the lumber, I focused on OPS+ and WAR Batting Runs (Rbat).

For ages 26-31 (his entire MLB career), Hamilton has a 135 OPS+ and 128 Rbat in 3,151 PAs. His OPS+ has been pretty consistent, with four out of six years falling between 131 and 139, with one year at 170 and another at 90.


The Comparison Group

I formed a comparison group as follows:*

  • For the years 1920-2007;
  • Ages 26-31 (or start of career through age 31);
  • OPS+ from 130 to 140; and
  • 2,500+ PAs.

There are 98 players in this group. For their years played at age 26-31, the group averaged 137 games (Hamilton 123), 577 PAs (Hamilton 525), 134 OPS+ (Hamilton 135) and 23.6 Rbat (Hamilton 21.3). Per 150 games, the group averaged 25.8 Rbat, Hamilton 26.1.

Now let’s see the group’s performance for age 32-36. Almost two-thirds (63 of 98) were active each year from 32-36, and the group averaged 4.30 seasons in that range. Only Mike Epstein and Whitey Kurowski were finished by 31.


The Basic Comparison

Here are the group averages, first for age 26-31, then for 32-36 based on potential years played (figures in parentheses are based on actual years played), and the net change:

  • Games/year: 137 … 100 (116) … Minus 27% (minus 15%)
  • OPS+ (weighted): 134 … 121 … Minus 10%
  • Rbat/year: 23.4 … 10.6 (12.4) … Minus 55% (minus 47%)
  • Rbat/150 games: 25.6 … 15.9 (16.0) … Minus 38% (minus 37%)



More than half of this group mainly played outfield in the 26-31 period (52 of 98), so let’s examine them separately. To my surprise, the OF subset was not significantly more durable or productive at 32-36 than the rest of the group. Here are the averages for the OF subset, first for age 26-31, then for 32-36 based on potential years played (figures in parentheses are based on actual years played), and the net change:

  • Games/year: 136 … 101 (116) … Minus 26% (minus 15%)
  • OPS+ (weighted): 138 … 123 … Minus 11%
  • Rbat/year: 23.9 … 11.6 (13.2) … Minus 51% (minus 45%)
  • Rbat/150 games: 26.4 … 17.2 (17.1) … Minus 35% (minus 35%)


Power Hitters

Hamilton has averaged 33 HRs per 650 PAs. Of the comparison group, 52 of 98 averaged at least 25 HRs per 650 PAs for age 26-31. Did this subset retain more or less of its offensive value than the overall group? Here are the averages for the power-hitting subset, same format as before:

  • Games/year: 137 … 96 (114) … Minus 30% (minus 17%)
  • OPS+ (weighted): 134 … 119 … Minus 11%
  • Rbat/year: 23.1 … 8.5 (10.1) … Minus 63% (minus 56%)
  • Rbat/150 games: 25.3 … 13.3 (13.3) … Minus 47% (minus 47%)

The power hitters lost significantly more of their output than the overall group, whether measured per year or per 150 games.

Strong Years

How many strong years did the full group produce during age 32-36? Since they averaged 23.4 Rbat for age 26-31, I calculated the percentage of years with at least 23 Rbat for each period:

  • Age 26-31: 54% of years had 23+ Rbat.
  • Age 32-36: 21% of potential years had 23+ Rbat (24% of actual years).

For age 32-36, the number of players with exactly N years of 23+ Rbat (number with N or more such years in parentheses):

  • 5 years — 4
  • 4 years — 4 (8 with 4 or more)
  • 3 years — 8 (16 with 3 or more)
  • 2 years — 5 (21 with 2 or more)
  • 1 year — 32 (53 with 1 or more)

That’s pretty stark, no? Almost half the players in this group (45 of 98) never produced 23+ Rbat in a season from 32-36. And 79% produced no more than one such year.

(By the way, the four who reached 23+ Rbat in all five years age 32-36 were Edgar Martinez, Roberto Clemente, Bob Johnson and Charlie Gehringer. For total Rbat age 32-36, Edgar stands head, shoulders and ribcage above everyone else in this group, producing 284 Rbat — 50% more than #2 (Clemente). Only three men in MLB history produced more Rbat for 32-36 than Edgar: Babe Ruth (449), Barry Bonds (323) and Mark McGwire (300). Only 17 totaled 200+ Rbat for that age range.)


In Conclusion

The median Rbat for the full group at age 32-36 combined was 40, or 8.0 Rbat per year. Out of 98 players, the top 18 produced at least 100 Rbat for 32-36, but there’s a sharp drop from there; #25 produced just 71 Rbat, and #30 had just 60.

Given Hamilton’s injury history and other risk factors, the most reasonable expectation would be that his age 32-36 production does not exceed the modest average of the group.



* I made a cursory effort to prevent WWII stats from distorting the study, but I wound up excluding only one player who met the group criteria: Nick Etten had four of his five good MLB years during the talent shortage of 1942-45, with only 1,300 PAs outside of those years; that’s just not enough to judge his true MLB ability. A few other wartime stars were left in because they had multiple good years outside of wartime, such as Bill Nicholson, Phil Cavarretta, Roy Cullenbine, Bob Johnson, Lou Boudreau, Stan Spence and Whitey Kurowski.

Spence, for example, averaged a 138 OPS+ from 1942-44, missed ’45, then averaged that same 138 while playing full-time in 1946-47. He hit well in the top minors for several years just before the war. I just don’t think that the war years — his age 27-29 — substantially distort the picture of his ability.

I wavered on Kurowski, whose first full year was ’42 and who had just 1,500 PAs outside of wartime. I ultimately left him in because his best two-year span came after the war, and it’s clear that postwar injuries (and not the leagues’ return to normal talent levels) led to his early demise.

0 0 votes
Article Rating
Notify of
Newest Most Voted
Inline Feedbacks
View all comments
9 years ago

57 players have totaled between 22.3 WAR and 24.3 WAR over their age 26-31 seasons (within one of Josh’s total). Of those 57 players, the most WAR produced over their age 32-36 seasons has been: Dwight Evans and Art Fletcher, 20.8 each Willie Stargell 20.1 Derek Jeter 19.8 Stan Hack 18.8 Dave Winfield 17.8 Of that same group of 57, the players with the least productive age 32-36 seasons(or earlier, for careers that ended before age 36): Curt Flood comes in lowest at -0.4, but for righteous reasons, unrelated to the playing field After Flood, there’s Mo Vaughn 0.0 Garry… Read more »

9 years ago

I wonder if there is a universal human tendency to forget on purpose the inevitable ravages of time on our minds on our bodies? Rational behavior on the part of sports teams would be to cut all their stars adrift in their early thirties — no certainty, but odds working strongly in your favor — but they don’t – CEOs should be retired at 55 but they don’t.
Members of congress….

Jason Z
9 years ago

I think this iformation fits neatly with previous discussions we
have had defining a players prime.

Not to analyze the financial side per your wishes John, but suffice it
to say that if you sign a player to a multi-year deal, when said player
is past 30, enjoy compensating that player for what he did previously.

This may not always be true, but will be true enough times to make it
bad business to even attempt.

Jason Z
9 years ago

I know that John. What’s to analyze? The Dodgers and Angels
are in a no holds barred spending orgy of stupidity to try
and grab the SOCAL baseball fan.

If some huge earthquake doesn’t cause California to break off
into the Pacific, all these huge contracts may just do the trick.

I wonder how many examples of ill advised contracts baseball execs.
need to see before this changes. I guess there is always someone who
thinks it will work for them.

9 years ago
Reply to  Jason Z

Here’s a table showing teams since 2001 who had 4 or more batters with 400+ PAs, either aged 31 or younger, or aged 32 or older. Age # of Teams 2+ WAR 2+ WAR 3+ WAR 3+ WAR up to 31 283 113 39.9% 34 12.0% 32 and older 47 7 14.9% 1 2.1%   So, it’s really only a small number of teams that have tried to load up on aging players, and most have had mediocre results. Many more teams have tried to do the opposite (because it’s a lot less expensive), and have had rather more success… Read more »

Mike L
Mike L
9 years ago

Economically, we may be looking at it incorrectly (and the Yankees may be smarter this year than they seem). I really doubt that any team signing a player to a enormous long term contract for someone who in already in his thirties really expects to see much value at the back end of the deal. What they really are doing is overpaying for the front end, hoping there are no early injuries and they luck out and have a moderately productive player at the end. Hamilton (and Greinke) might meet those expectations. I kind of doubt it.

9 years ago
Reply to  Mike L

I agree, Mike. But, the odds are still not great.

John’s original comparison group of 98 breaks down almost exactly into thirds in terms of either 0, 1 or 2+ good seasons (125 OPS+ in 400+ PAs) aged 32-37.
– ZERO seasons, 34 players
– ONE season, 31 players
– TWO or more seasons, 33 players
– THREE or more seasons, 19 players (~20%)

Jim Bouldin
Jim Bouldin
9 years ago

Off topic*: I’m wondering if there’s going to be a HHS “year in review” type thing at any point.

*and a plug for the utility of an open thread here….

Timmy Pea
Timmy Pea
9 years ago

Hamilton is flaky, and prone to personal failings, he is a risk. I wonder which way he’ll go? Stay in great shape hoping for more cash after this contract or will he lose enthusiasm. I’ve watched him a lot and the guy can practically phone in 30 HR’s.

Jim Bouldin
Jim Bouldin
9 years ago

I’m doing some simulations of offensive strategy and I need some data on runner success rates to help inform model parameter estimates, if anybody has the inclination and can help me out. Specifically, I need info on (1) success rates in going from first to home on doubles and first to third and second to home on singles (2) wild pitch, past ball, balk and pickoff rates, (3) attempt and success rates in stealing second and third base. If at all possible I’d like the data for both aggressive and non-aggressive base-running teams. Double play and hit and run data… Read more »

9 years ago
Reply to  Jim Bouldin

Jim, on Baseball-Reference you can find seasonal info on (1) and (3) under Batting/Detailed Stats on a team’s page, or MLB-wide under Batting/Baserunning. For (2) you should probably be able to find it under Pitching/Detailed Stats for teams. The info should also be there on the MLB-wide page, but I’m not looking right now because the sheer amount of information on each page is crashing my not-updated-in-several-years work browser. Here’s what I’m talking about:


Hope that helps.

Jim Bouldin
Jim Bouldin
9 years ago
Reply to  RJ

Helpful indeed, thanks for the pointers RJ.