Percentage of full seasons played by rookies

Here’s a simple calculation that reveals lots of interesting stuff. Using’s new option to limit searches to players who qualify as rookies (using the modern definition only), it’s easy to find what fraction of players each season who qualified for the batting title also qualified as rookies.

This is what the data looks like:

percentage rookies

The overall trend is clear. In the early part of the 20th century, about 15% of full-time players were rookies. There is an uptick in 1914 as the Federal League was formed and lots of new players were brought in. In the late 1910s, nearly no rookies played full seasons but by the 1920s the rate was back up to about 12%. When World War II came around, the rate spiked as a lot of established players left for military service. There was a spike from 1952 into 1953, due to the Korean War. The rate spiked even further in 1954, but I’m not sure why, as that war was already over and guys like Ted Williams and Willy Mays were back.

By the 1960s, the rate of rookies dropped down to about 9%. In the 1970s, it went even lower, plummeting to an all-time low in 1979. From 1980 on, the rate has been relatively consistent around 5%. Note a few upticks in expansion years of 1969, 1977, and 1993.

I wonder why the fraction of rookies has dropped gradually over time? September call-ups probably have something to do with it–more players qualifying as rookies before playing a full qualifying season. There may just also be less playing of rookies full time–for example fewer rookies winning jobs in spring training.

What else can we surmise from this data?

16 thoughts on “Percentage of full seasons played by rookies

  1. 1
    Mike L says:

    Service time and its effect on arbitration might play a role. Also multiyear contracts for veterans which locks the team in place and the rookie in triple a. And perhaps free agency itself. In the past if you needed a player you either had to trade for him, perhaps creating another hole, or bring up whomever you had in the minors. Now you can just buy him in the offseason. The 54 spike and the echo over the following two years is interesting. I wonder if it somehow correlates in age to retirements of those who came back from WWII.

    • 2
      Richard Chester says:

      In 1954 10 of the 13 qualifying rookies were on just 4 teams: 3 each on the A’s and Pirates and 2 each on the Cubs and Tigers.

  2. 3
    brp says:

    I’m not sure how big the effect of either one is, but here’s two thoughts:

    1) DH rule in 1973 has allowed some players to stay around in the AL longer than they might have otherwise. I realize we’re talking at most only a dozen guys or fewer, but those are roster spots that can’t go to a rookie then.
    2) Better training and nutrition allows for longer careers for some players, I’m sure; and that dip around 2002 meshes pretty well with the steroid era.

    I think both of those combined is probably just a 1-2% at most, though, when compared to say the 1950s.

    The “bonus baby” rule from 1947-1965 may be a slight factor as well, though I think the vast majority of those players didn’t play enough to have qualifying seasons.

  3. 4
    Doug says:

    Just a guess, but I suspect that a plot of the average age of those rookies overlaid on the chart would show an opposite trend towards older rookies. That could explain why the % of rookies has declined – they’re kept in the minors longer to make sure they are ready when they’re called up.

    Another factor is free agency. Before UFAs, if you had a hole to fill in your lineup and you couldn’t swing a desirable trade, you pretty much had to bring up someone from the farm (could be a kid, or a re-tread vet). Now, you just grab a UFA.

  4. 5

    It looks like 1979 is below 1%. Does that mean two rookies qualified for the batting title that year? 26 teams * 8 hitters = 208 qualifiers, which is probably on the high end given platooning/injuries.

  5. 6
    mosc says:

    I might lower the limit rather than batting crown eligibility to something more reasonable to accumulate significant single season value. I feel like the rate for rookies contributing in their official “rookie” capacity has more sharply declined over time. It might be interesting to look at total WAR by “modern rookies” over the years too, which should also dip. Although that one might have an odd 2012 spike in it…

    I sense this coming out of the discussion on rookies sort of wasting their rookieness on fractional seasons but not sure of a better way to look at it. Another thought that comes to mind is the percentage of career WAR in the rookie season through the years. Just throwing out ideas…

  6. 7
    Hartvig says:

    I don’t have time this morning to check this out but a few thoughts:
    1) you mentioned the correlation of some of the spikes & the Federal League & WW2 & Korea. They also seemed to have happened during expansion seasons (or did you mention that and I missed it?)

    2) Bill James wrote something about the number of minor league games played and how it relates over time- the effect of outside events like war and of things like expansion is obvious but I recall him saying that he couldn’t find any evidence that under normal circumstances players were being “seasoned” in the minors any longer than they used to be.

    3) One other thought was an article that Stephen Jay Gould (the Harvard paleontologist) wrote about how as species as a whole mature over time (in this case baseball is the species) you tend to find fewer “outliers”, that is members of the species that differ radically from the norm. I do think you tend to see fewer really young players- the last 17 year old I can remember was Young way back in the 70’s. I don’t think that’s true on the other end of the scale- I think they are at least as many if not more “old” players around- but that’s probably due to factors like better training and conditioning plus more time to do those things because most players no long need off-season jobs plus the financial incentives for playing longer are much greater now.

    • 8
      Andy says:

      Your 3rd point is a really good one. In general, scouting and minor-league training have become much more uniform over the years, to where outliers have become much less rare.

  7. 11
    John Autin says:

    Hartvig’s point #3 is of a piece with a long essay in the BJHBA about evaluating the competitive level of a league. He called it “Peripheral Quality Indicators” and it’s in the Bob Lemon player rating.

    Bill’s #2 PQI was the average distance of the players from age 27. And I think the decline in the rate of rookies who play regularly reflects the general rise in the level of play.

    I do think that in the last 10 years or so, teams controlling service time for arbitration reasons plays a big role, as Mike noted. But the biggest factor, I think, is just the size of the talent pool. We now have not quite twice as many MLB teams as there were up to 1960, but they draw from a pool that’s way more than twice as big — because of population growth, integration and international scouting & development — and MLB teams are far more sophisticated in finding that talent.

    When I read old histories, it’s amazing how many MLB players got their first pro job just because they knew someone on a local team who told them about a job opening. The process of finding the talent has improved immensely.

    And, there aren’t as many chronically awful teams nowadays. I think if we broke down the qualifying rookies, we’d find a disproportionate number of them on teams during long down phases, like the Phillies, A’s, Browns, Braves, etc. While KC and Pittsburgh fans might dispute the point, we just don’t have teams passing entire decades as doormats any more.

    • 12
      John Autin says:

      For 1901-20, the number of qualifying rookies for some teams:

      Worst teams (during the period):
      33 – Cardinals
      24 – Browns
      24 – Braves
      23 – Dodgers
      21 – Yankees
      19 – Senators

      Best teams:
      6 – Cubs
      9 – Giants
      12 – White Sox
      12 – Tigers
      14 – Red Sox
      23 – Pirates

      Only Pittsburgh doesn’t really fit the pattern that the weaker teams use more rookies.

    • 13
      Hartvig says:

      “Hartvig’s point #3 is of a piece with a long essay in the BJHBA about evaluating the competitive level of a league”

      And you’re right as always, of course.

      But I’m certain there’s a connection somewhere to Gould with the article. I’ll have to see if I can find my book by him that’s got the baseball stuff in it and see if that’s where it is. He may have applied James’ line of reasoning to the maturation of a species rather than the other way around.

      • 14
        Brendan Bingham says:

        Hartvig: I can confirm your SJ Gould memory, although I cannot quite nail down the source.
        I once read an essay of his in which IIRC he was talking about biological outliers. Rapidly evolving species (or perhaps populations living in a rapidly changing ecosystem) produce a wider range of biological forms than well-established species in stable environments. He drew the analogy to baseball, as he sometimes did in such essays, that pitchers winning 30 games in a season and batters hitting .400 used to be much more frequent. Such accomplishments were always in the tail of the distribution, even in the rapidly emerging “ecosystem” of early 20th century baseball. But with time came maturity and consistency which has tended to limit variation in performance, such that in the modern game feats like winning 30 games and batting .400 are now vanishingly rare.
        I read this essay as a student in about 1985, and I think it was either in Natural History magazine, for which Gould did a monthly column for many years, or in one of the many collections of Gould’s essays (from NH and other sources). That I read it in 1985 is not to say that it was written in 1985; it might have been from much earlier.

        • 15
          Lawrence Azrin says:

          Yes, I also recall Bill James referencing SJ Gould’s theories on biological outliers and the maturation of a species, I think it was in BJ’s book on the HOF.

          James’ point was that the average level of talent in MLB in the late nineteenth/early twentieth century was much lower, so that the very greatest players could exceed that lower league average by a lot more than recent/current MLB players could. So, for example, Rogers Hornsby’s .424 BA in 1924 is not THAT much better than George Brett’s .390 in 1980 or Nomah’s .372 in 2000.

          Another great point he made was that MLB talent is NOT equally distributed; we are analyzing the extreme right-hand size of the talent spectrum, and outliers will occur at irregular intervals.

  8. 16
    Ed says:

    Since we’re looking at percentages and not raw data, it’s impossible to know if this trend is due to 1) fewer rookies meeting the threshold, 2) more veterans meeting the threshold, or 3) a combination of both. I’ll offer a few possibilities why there may be more veterans meeting the threshold:

    1) Fewer roster spots for position players. This of course leads to less platooning, fewer days off in general, etc.

    2) Quicker recovery time from injuries. An injury that might have sidelined someone for 6-8 weeks just a few years ago, now leads to someone only missing 2-3 weeks. And hence they’re more likely to meet the playing time threshold.

Leave a Reply

Your email address will not be published. Required fields are marked *