Rogers That: Hornsby’s Prowess Leads Me to Revise My Pick for the Greatest 2B Ever

A Cousin’s Reference to Hornsby’s Bat as “24-Carat” Prompts an Analysis of History’s Greatest Second Basemen

My first introduction to the world of sabermetrics came at the tender age of 14.  Early in the summer of 1993, I purchased The Bill James Historical Baseball Abstract (2nd Edition), then pored over the text with two like-minded cousins, Jerrett Andrew and Ryan Facer.  The second edition of James’ classic work included the principles of peak and career value, powerful notions that influence my player analysis to this today.  (Sadly, later editions of James’ Historical Abstract did away with this concept, making way for the flawed, extensively-criticized Win Shares rating system the author presumably developed in response to single-number rating systems such as WAR and Total Player Rating.)

Ryne Sandberg had long been my favorite player of all time, so James’ ratings of second basemen were of particular interest to me.  In particular, his valuation of Joe Morgan as the game’s greatest second baseman in terms of peak value resonated heavily with me.  After all, James noted, Morgan posted MVP-caliber seasons toiling in the Astrodome–and, later, in Cincinnati’s symmetrical concrete monstrosity, Riverfront Stadium.  A glimpse at Morgan’s 1976 season reinforces James’ assertion that Morgan’s peak value far surpassed any other second baseman in baseball history:

 .320  .444  .576  186  27 60 86.9%   184 .865  9.6

Incredible!  A second baseman who led the league in slugging percentage, OPS+, wRC, Offensive Winning Percentage, and WAR?  Toss in his slightly above-average performance on defense and you’ve got, in terms of single-season performance, one of the most valuable players in baseball history.

Ultimately, however, there is a second baseman who makes Morgan’s contributions seem meager by comparison.  This particular player, a notorious troublemaker wherever he went, someone who Bill James claimed couldn’t backpedal on pop flies, translating into an inordinate number of bloop singles for his opposition–and yet, simultaneously, a player who led the league in Wins Above Replacement an astonishing seven different times!  I am referring, of course, to the legendary–perhaps notorious–Rogers Hornsby.

While Hornsby unquestionably benefited from the lively-ball era ushered in by Babe Ruth in 1920, the truly important metrics–WAR, WPA, OPS+, and adjusted Batting Runs are all adjusted to compensate for the era in which Hornsby played.  Nevertheless, it is jaw-dropping to view Hornsby’s raw numbers during the roaring ’20s;

  • Batting averages of .387, .397, .401, .403, and .424
  • Six (!) consecutive seasons leading the league in on-base percentage, slugging percentage, and OPS+
  • Eight (!) seasons batting .370 or better
  • A 1922 statistical line of 250 hits, 46 doubles, 42 home runs, 152 RBI, a .401 batting average, and 450 total bases

Only Babe Ruth and Barry Bonds can boast such consistently astronomical statistical lines.  And yet, be it his reputation for lackluster defense, his clubhouse troublemaking, or his existence in the shadow of Ruth, Hornsby is rarely included in discussions of the game’s greatest all-time players, even when the discussion is narrowed to just second baseman.  Indeed, if I recall correctly, James had both Morgan and Jackie Robinson ahead of Hornsby in terms of peak value, the latter an utterly unjustifiable selection likely made as part of James’ agenda of political correctness.

(After all, in his most recent installment of the Historical Baseball Abstract, James selected Oscar Charleston as the third-best player to ever suit up, despite the fact that he never earned a single win share or faced Major League competition.  Selecting Charleston ahead of Willie Mays, Mickey Mantle, Lou Gehrig, Barry Bonds, Ted Williams, Stan Musial, Hank Aaron, and many, many others may only be interpreted as James’ desperate attempt as a white liberal to include Negro League players despite the fact that the author’s entire book was based on rating players based on their statistical records.)

James’ maniacal liberalism aside, a closer look at Hornsby’s peak seasons reveals just how dominant a second baseman he was in the National League of the 1920s:

 1917  41.3  164  169  .796 9.9
 1921  78.3 191  191 .845  10.8
 1922  96.3 198   207 .864   10.0
 1924  97.8 221   222 .895  12.1
 1925  87.5 208   210 .883 10.2
 1927 63.8 177   175  .812 10.1
 1929  75.9 174   178  .823  10.4

Hornsby’s reign at the plate during the ’20s was so complete that the two seasons not included in the table above–1920 and 1923–saw the legendary second-baseman hit .370 and .384, respectively.  Of course, astronomical batting averages weren’t necessarily uncommon during Hornsby’s era, but you nevertheless didn’t find many players, regardless of position, who posted OPS+ figures of 185 (1920) and 187 (1923) in their off-seasons.

As previously mentioned, I long held that Joe Morgan was baseball’s greatest second-baseman, in terms of peak value.  Later, I came to prefer Eddie Collins, who, despite hitting for virtually no power, got on base so often for so many years–and wasn’t shabby with the glove in his hand, either–that he, too, seemed a potentially worthy candidate.  For the purposes of this exercise, I thought it useful to examine the individual components of Wins Above Replacement (WAR) for Morgan, Hornsby, and Collins, for each players’ five best seasons and their respective careers, and see if doing so gets us any closer to a truly definitive answer.

Before examining the seasons and career of these legendary players, however, it is necessary to explain the terms that factor into the WAR figure we’ve become so familiar with as a litmus stick for our favorite baseball stars:

Rbat: Number of runs better or worse than average the player was as a hitter
Rbaser: Number of runs better or worse than average the player was for all base-running events (SB, CS, PB, WP, defensive indifference)
Rdp: Number of runs better or worse than average the player was at avoiding grounding into double plays
Rfield: Number of runs better or worse than average the player was for all fielding events
Rpos: Runs from positional scarcity
WAR: Total Wins Above Replacement


YEAR Rbat Rbaser Rdp Rfield Rpos WAR
1972 42 8 3 7 4 9.3
1973 43 7 1 11 5 9.2
1974 46 8 1 4 4 8.6
1975 55 10 2 14 4 10.9
1976 57 10 2 -1 4 9.6
22 Seasons 450 80 25 -48 73 100.3


YEAR Rbat Rbaser Rdp Rfield Rpos WAR
1909 45 5 0 11 0 9.7
1910 33 10 0 24 0 10.5
1911 45 -1 0 -1 0 6.5
1912 48 7 0 6 0 8.8
1913 46 2 0 11 0 9.0
25 Seasons 628 40 0 35 37 123.9


YEAR Rbat Rbaser Rdp Rfield Rpos WAR
1921 76 -0 0 10 3 10.8
1922 95 0 0 -13 5 10.0
1923 54 -1 0 1 2 6.7
1924 96 -2 0 7 5 12.1
1925 88 1 0 0 5 10.2
23 Seasons 861 -9 0 54 75 127.0


If there was ever a case of the numbers reinforcing what we believe to be true, this is it.  Hornsby was a monster with the bat in his hands, Collins was the steadiest on defense, and a huge part of Morgan’s value came from the havoc he created on the basepaths.  Unsurprising, too, is that Hornsby wasn’t nearly the liability at second base that James, et al painted him to be over the years.  Morgan’s defensive totals don’t align that closely with his reputation as a quality fielder, though -41 of his career -48 Rfield were accumulated over the final eight seasons of his career, when his range wasn’t nearly what it once was.

Conclusions?  Well, Hornsby has the single-greatest season in terms of WAR (12.1; 1924) and the highest career total (127.0) despite having roughly 3,000 fewer plate appearances than Collins.  Can we, finally, once and for all, declare “Rajah” the greatest second-baseman to play the game?  It might not be as iron-clad as us absolutists would like, but, yes, I think the time is finally here.


Ron Santo: Baseball’s Best 3B in the 1960s

Statistical Analysis Demonstrates that Santo, and Not the “Human Vacuum Cleaner,” Was Baseball’s Best at the Hot Corner During the ’60s

Ron Santo was unquestionably the best National League third-baseman during the decade of peace and love.  But was Santo truly a better player than Baltimore’s Brooks Robinson, the third-baseman that defined the position for millions of Americans during the 1960s?  After all, Robinson was a first-ballot selection to the Baseball Hall of Fame, while it took years of campaigning to earn Santo his rightful place in Cooperstown–and this honor, sadly, was bestowed upon Santo after the legendary Cubs third-baseman and broadcaster had passed away after years fighting diabetes.

The argument that Santo was indeed the greater overall player must be approached using two distinct methods of player-value analysis.  Firstly, we will compare each player’s five best consecutive seasons to measure the third-basemen’s peak value.  We will then measure their total contributions over the entire decade to assess career value–though, in both cases, Santo and Robinson continued playing well into the next decade.


Santo, of course, played for some mediocre Chicago Cubs  teams.  That said, he was an offensive powerhouse during one of the most pitching-dominant eras in baseball history.  Over the course of the decade, Santo led the National League in walks 4 times, on-base percentage twice, triples once, and also posted the best Wins Above Replacement figure (9.8) in 1967, meaning he was the most valuable player in the National League that season–better than Willie Mays, Roberto Clemente, Dick Allen, and the rest of the decade’s superstars.

Santo’s five best consecutive seasons–logged from 1963-67–are summarized below:

1963 .297 .339 .481 128 25 99 42 129 .659 1.7 6.7
1964 .313 .398 .564 164 30 114 86 164 .791 5.8 8.9
1965 .285 .378 .510 146 33 101 88 145 .736 4.9 7.7
1966 .312 .412 .538 161 30 94 95 157 .771 5.1 8.9
1967 .300 .395 .512 153 31 98 96 153 .753 3.8 9.8

Of note during Santo’s 1967 campaign is that 2.7 of his Wins Above Replacement total came courtesy of the third-baseman’s contributions with his glove.  In fact, according to the WAR system, Santo was the second-best defensive player, regardless of position, in the National League in 1967.  Of Santo’s five peak seasons outlined above, 1963 was the only other campaign in which Santo finished as one of the top-10 defensive players in the Senior Circuit based on WAR.  That said, he was one of the league’s top-10 offensive contributors all five seasons between 1963-67.

Santo’s defensive value is of particular importance, since, in this study, he is being evaluated in comparison to Brooks Robinson, whose glove work earned the Orioles legend an astounding 16 Gold Glove awards over the course of his career.  As luck would have it, Robinson’s best five consecutive seasons were 1964-68, meaning that, aside from obvious differences resulting from Robinson having played in the American League during that period while also calling Memorial Stadium his home for 81 games each season, we are able to compare the two players over virtually the same period in time.  Besides, complex sabermetric measures such as OPS+, WPA, and WAR factor in differences in league, home park, and era, allowing us to view each player’s peak side by side without worrying about the distorting effects often manifested by differing leagues and home stadiums.

Below are the records of Robinson’s best five consecutive seasons from 1964-68:

1964 .317 .368 .521 145 28 118 51 145 .706 4.8 8.1
1965 .297 .351 .445 124 18 80 47 126 .644 1.1 4.5
1966 .269 .333 .444 123 23 100 56 121 .607 3.1 4.6
1967 .269 .328 .434 124 22 77 54 123 .598 -0.1 7.7
1968 .253 .304 .416 117 17 75 44 116 .589 2.0 8.4

At first glance, the stark difference between Santo and Robinson couldn’t be any more evident.  During the five-year peak periods of each respective player, Santo

  • Out-homered Robinson 149-108;
  • Logged four seasons with an on-base percentage better than Robinson’s best single-season performance;
  • Posted four seasons with an OPS+ better than Robinson’s highest mark of 145 (logged in 1964);
  • Drew 407 bases on balls compared to Robinson’s 252;
  • Created 748 wRC (weighted Runs Created) versus Robinson’s 631; and,
  • Posted four seasons with an Offensive Winning Percentage better than Robinson’s best single-season mark (1964)

Even more damning, despite clearly being outperformed at the plate by Santo, Robinson consumed more outs (2,309, compared to 2,203 for Santo) in the process.  Ultimately, Robinson performed at a lesser level than Santo in the batter’s box, yet cost his team nearly 5 percent more outs along the way.)

For as thoroughly as Santo outclassed Robinson at the plate during the ’60s, the WAR totals of the two Hall of Fame hot-corner men aren’t as disparate as you’d otherwise suspect.  To be sure, Santo holds a commanding lead–42 Wins Above Replacement in five seasons versus Robinson’s 33.3–and yet, given how complete Santo’s dominance was with a bat in his hands, one might expect the gap to be even greater.  There’s a very simple reason for this: the Wins Above Replacement metric includes offensive, base-running, and defensive contributions–and, as we’ve known all along, Robinson’s claim to being the best third-baseman of the decade has always hinged on his glovework being vastly superior to Santo’s or anyone else.

As most baseball fans know, defensive statistics have long stood as an impenetrable labyrinth from which very little of analytical value could be absorbed.  Over the past 20 years, however, advances in defensive metrics have made it possible to reliably evaluate any individual player’s contributions with the glove beyond rudimentary measures such as fielding percentage and the number of Gold Glove awards won.  While such defensive analysis is still in its infancy, there now exist useful metrics we may employ to evaluate the respective defensive contributions of Santo and Robinson.

For grins, let’s begin with the number of Gold Glove awards each player collected during his respective five-year peak during the ’60s.  Ron Santo earned four Gold Gloves (1964-67), while Brooks Robinson went five-for-five between 1964-68, and added a Most Valuable Player trophy (1964) to boot.  Needless to say, it is clear that both players’ contemporaries viewed them as master glovemen and better at fielding the third-base position than anyone else in the National and American leagues.

As a crude baseline, let’s also look at each player’s fielding percentages during their respective five-year peaks:

Ron Santo Brooks Robinson
.951 (1963) .972 (1964)
.963 (1964) .967 (1965)
.957 (1965) .976 (1966)
.956 (1966) .980 (1967)
.957 (1967) .970 (1968)

Needless to say, Robinson fielded his position with much more accuracy than Santo.  That said, Santo saw considerably more total chances fielding third-base for the Cubs–his total for the five peak seasons outlined above is 2,803.  Robinson, meanwhile, handled a total of 2,548 chances at the hot corner in Baltimore, or roughly 1% less than his Chicago counterpart.

Admittedly, these rudimentary statistics provide only a limited glimpse into the true fielding prowess of Ron Santo and Brooks Robinson.  Ultimately, what we are concerned with is how many runs above average each player saved his team through his glovework at the hot corner.  Using the Ultimate Zone Rating (UZR) method of calculating runs saved by defensive players, we are able to see the contributions of both Santo and Robinson at third-base during the ’60s:

Ron Santo Brooks Robinson
10.0 (1963) 17.0 (1964)
3.0 (1964) 8.0(1965)
7.0 (1965) 4.0 (1966)
7.0 (1966) 32.0 (1967)
18.0 (1967) 33.0 (1968)

While Santo’s performance at third-base certainly reinforces his selection as a Gold Glove award winner four of the five seasons included in the peak of his career, Robinson’s totals serve to validate his reputation as the greatest fielding third-baseman–not just during the ’60s, but perhaps of all time.  (Indeed, Robinson continued his stellar glovework through 1975, logging runs-saved totals between 15-23 [with one exception] through 1975.)

Despite Robinson’s obvious advantage on defense, we must remember that these runs-saved figures factor into the calculation of Wins Above Replacement.  As mentioned previously, Santo’s WAR total over his five peak seasons was 42, a substantial edge over Robinson’s 33.3.  In other words, when offensive, base-running, and defensive contributions are totaled, Ron Santo clearly emerges as the greatest third-baseman of the ’60s based on peak value.


While Santo earns the title of the best third-baseman of the 1960s in terms of peak value, one question remains: Which player posted the better career value over the course of the entire decade?  It makes sense to begin with Wins Above Replacement, which, for better or worse, is the final arbiter of player value both for a single season and over the course of a career.

Ultimately, Santo and Robinson are virtually equal in terms of WAR, with Santo posting 56.3 Wins Above Replacement from 1960-69, while Robinson logged a WAR total of 54.5.  Santo’s total is rendered even more impressive considering he played in only 95 games in 1960, while Robinson was an everyday player each season between 1960-69.

Considering Santo’s slightly-reduced playing time and his contribution of nearly 2 more Wins Above Replacement than Robinson over the course of the decade–and, considering that Santo was clearly the more valuable player in terms of peak value during the ’60s–the only conclusion to be reached is that, indeed, Ron Santo is the greatest third-baseman of the peace and love decade.

Of course, in terms of total career value, Robinson has the obvious edge, as he played far longer than Santo and earned more Wins Above Replacement over the course of his career.  But total career value is not what this analysis is about; rather, we set out to identify the greatest third-baseman of the ’60s, and we’ve found that answer.  Congratulations to Ron Santo, an overlooked, underrated, and often forgotten all-time great that toiled for awful Cubs teams during the 1960s.  You ultimately earned your spot in Cooperstown, even though it took far, far longer than it should have to get you enshrined.

[Edits: Added sub-headline, corrected minor spelling errors, and provided links to glossary entries explaining some of the more sophisticated metrics.]