April 29, 2007
The "Deadball Era" stretched from 1901-1919, and it was a time when batters stretched double into triples, and triples into inside-the-park home runs more often than they hit balls over the fence. Teams bunted. They utilized the hit-and-run. They stole a lot of bases. It was a style far different than what many of today's statistically-savvy fans consider "smart baseball." However, given the many factors that generated that style of play, was it smart baseball in its time?
David talked to baseball historian Dan Levitt, the co-author (with Mark Armour) of Paths to Glory, and a contributing author to SABR'S Deadball Stars of the American League and Deadball Stars of the National League, about some of the era's best players, and about how the game was played.
Baseball Prospectus: Who were the sabermetric stars of the Deadball Era, and who were the unsung stars?
Dan Levitt: The top players of the time, as evaluated by the sabermetric stats, are generally the same as those traditionally recognized as the best. I listed two second basemen because they were close in value and head-and-shoulders above the rest of field. Note, I considered mainly the player's contribution during the deadball era, usually defined as 1901-1919:
C Roger Bresnahan 1B Frank Chance 2B Eddie Collins 2B Nap Lajoie 3B Frank Baker SS Honus Wagner OF Ty Cobb OF Joe Jackson OF Tris Speaker P Walter Johnson P Pete Alexander P Christy Mathewson P Mordecai Brown
As to some stars that may be unsung, Bresnahan was an excellent player who is sometimes wrongly dismissed as a Hall of Fame mistake. Giant Larry Doyle and Athletic Danny Murphy were both very good second basemen. Miller Huggins, who later gained fame as a Yankee manager, was a fine second baseman for the Cardinals. Another Giant, third baseman Heinie Groh, was an excellent player now remembered for little more than his "bottle bat." A third Giant, catcher Chief Myers--one can see why New York won so many pennants--was an excellent hitter. Cleveland shortstop Ray Chapman is naturally remembered for his tragic death from a beaning, but he was a great player. In general, history has looked more kindly on position players whose careers included a significant portion of the 1890s at one end or the 1920s on the other when the higher offensive levels helped boost rate and career stats.
Joe Wood was a great hurler whose pitching career was cut short by injury; Noodles Hahn was also an excellent pitcher whose career ended prematurely. The surly Carl Mays, who threw the pitch that fatally beaned Chapman, may be the best eligible pitcher not in the Hall of Fame.
BP: Who are some of the other players who maybe don't receive the respect they deserve?
DL: I'm not advocating for him here, but I'm a little surprised catcher Wally Schang never made it into the Hall through the Veterans Committee. A great hitter for a backstop, Schang played regularly on six pennant-winning teams in three different AL cities (including New York), and well into the lively-ball era of the 1920s. He was not a star defensively, but not a liability either.
Center fielder Happy Felsch, banned for his role in the 1919 World Series fix, was a great fielder. He was recognized as such at the time, and a sabermetric evaluation of the statistics confirms it. Offensively, he was very good as well--if he had played into the lively-ball era, he would today be recognized as one of the era's best center fielders.
Longtime NL right fielder John Titus might be the best player no one has ever heard of.
BP: How appreciated were the power hitters of the era, guys like Buck Freeman, Piano Legs Hickman, Gavy Cravath, and Home Run Baker? And how valuable were they?
DL: Freeman, Hickman, and Baker were all good hitters, but not really home run hitters. Exclusive of Freeman's 25 homers in 1899, the trio never hit more than 13 in a season. Baker led the league four years in a row without ever hitting more than 12. In fact, Baker gained his nickname for hitting two in the 1911 World Series, not for leading the league that year with 11. These modest home run totals, even if sufficient to lead the league, were generally recognized as not really enough to dramatically alter the player's value in comparison to his peers. At these low absolute totals, the difference in home runs between players was less significant than differences in other offensive statistics. Home runs aside, Baker was a great player, and recognized as such at the time.
As Mark Armour and I discussed in Paths to Glory, Clifford "Gavvy" Cravath was the first real slugger of modern baseball. He led the National League in home runs six times in the seven years between 1913 and 1919, and missed by only one in 1916. Cravath was no one-dimensional slugger--he could draw a walk and hit for average as well; in 1913 he nearly won the triple crown. He was probably the best position player in the National League during the teens.
The Phillies' home ballpark at that time, Baker Bowl, was long regarded as a hitters' haven due to its short right field fence, and Cravath's reputation is lessened by Baker Bowl's reputation as an extreme offensive ballpark. In fact, in the teens the Baker Bowl was not the hitters' paradise it would later become. Upon a closer look, Cravath's accomplishments still stand up to scrutiny. In the Phillies pennant-wining year of 1915 Cravath hit 24 home runs, the highest twentieth century season total prior to Babe Ruth. Some have discounted this accomplishment because he hit 19 of his home runs in Baker Bowl with its short right field and tall tin wall. To understand the value of those home runs, however, one must place them in context. Visiting teams in Baker Bowl hit a total of 18 home runs that year; in other words, in his home games, Cravath single-handedly outhomered all the opposing players.
BP: What was the ratio of triples to home runs, and were triples the "home runs of the Deadball Era"?
DL: As the name implies, in the Deadball Era home runs were much less frequent for several reasons. For one, the ball was replaced only sporadically, producing a squishy ball after being in play for a short period of time. The ball was further despoiled by the saliva and foreign substances often used by the pitcher. Cavernous stadiums also helped hold down home runs. Finally, because the sum of these factors all but eliminated the home run, the standard swing of the hitters was geared to line-drive base hits and not to hit the ball over the fence.
During the Deadball Era a home run was hit on roughly 0.5 percent of at-bats, and a triple on 1.4 percent, nearly three times as often. Furthermore, a significant minority of the home runs--as high as 105 out of 259 in 1909--were inside-the-park. In other words, in 1909 the sixteen teams averaged less than ten over-the-fence home runs each. Today home runs are hit on more than 3.0 percent of at bats and triples on only around 0.6 percent. In the Deadball Era triples and inside-the-park home runs were as much a result of power as speed as hitters pounded the ball into the farthest reaches of the big ballparks.
BP: What is known about the success rate of steal attempts in the era?
DL: Today most analysts believe the break-even rate for stolen bases is around 70 percent; if one is successful less often the failed attempts subtract more from the scoring of runs than the successful steals add. This is of course a generalization, and in individual instances the break-even rate can vary dramatically. In fact, the overall stolen base success rate in the major leagues over the past couple of years is right around the 70 percent mark.
At a time when fewer home runs made scoring from first less likely, taking a greater risk to get into scoring position was acceptable. Although I can't offer a definitive answer on how much less the break-even percentage might have been during the deadball era, the actual success rate was roughly 55 percent. Managers were much more willing to sacrifice an out to get the runner to second. Along the same lines, the strategic use of the sacrifice bunt was much more common at the time.
A look at the stolen base percentages of the best base stealers is instructive as well. The top recent stolen base leaders, such as Rickey Henderson, Vince Coleman, Joe Morgan, Willie Wilson, and Tim Raines, all had career stolen base percentages above 80 percent. Unfortunately, caught stealing information is available only for select years in the teens: 1912 and 1914-16 in the AL, and 1913 and 1915-16 in the NL. Nevertheless, we can gain some insight from this limited sample. Ty Cobb stole 260 bases and was caught 113 times, for a 70 percent success rate. Fellow AL speedster Eddie Collins stole 207 bases, and was thrown out 103 times, for a percentage of 67 percent. Max Carey, the senior circuit's top stolen-base threat, swiped 160 bases in the three Deadball years and was caught 53 times, a stolen base percentage of 75 percent.
BP: Looking at batting orders, did most teams prioritize having their best base-stealers in the leadoff position, or those most proficient at getting on base?
DL: It's hard to generalize over sixteen teams over nineteen years. Having made my disclaimer, I would suggest that teams looked for a good batting average, speed, and bat control. Walks were regarded as a function of the pitcher and not generally attributed to the batter. While not specifically focusing on walks, managers looked for "bat control," loosely defined as the ability to foul off pitches and not strike out. In many instances, the combination of a high batting average and bat control acted as a proxy for a solid on base average, even if not explicitly defined.
BP: Roy Thomas led the National League in walks seven times. With so much emphasis on "small ball," how much was he valued at the time?
DL: Thomas was recognized as a very good player, but more for his bat control than his walks. As noted above, bat control sometimes acted as an inadvertent euphemism for an ability to draw a walk. Of course, sometimes bat control could refer to a player with few strikeouts, but not necessarily the ability to draw a walk.
BP: According to the Bill James Historical Abstract, four of the best seven pitchers of all time played in the Deadball Era. What, if anything, does that tell us?
DL: As I discussed in an article in the 2000 Baseball Research Journal, pitch counts have increased dramatically since the Deadball Era. In comparing pitch count data from the 1919 World Series to the one in 1997, I found that modern pitchers throw about 1.36 as many pitches per inning as their Deadball counterparts. Much of the increase can be explained by the increase in walks and strikeouts, which result in longer pitch sequences per plate appearance.
Using the not-unreasonable assumption that pitch counts are the correct measure of starting pitcher workload, a Deadball Era pitcher could toss a much larger percentage of his team's innings than a modern hurler. Using the 1.36 multiplier suggests 225 innings today results in the same number of pitches--and thus the equivalent workload--as 306 in the Deadball Era. A cursory test suggests the multiplier is reasonable: in 1917 (1918 and 1919 were shortened seasons), the top five NL pitchers averaged 331 innings pitched; 1.33 more than the top five in 1997, who averaged 249.
James's quartet of Johnson, Mathewson, Alexander, and Cy Young were undoubtedly great pitchers, but in terms of value they benefited from a style of play that allowed them to participate in a greater percentage of their team's action. My point is not that these players do not deserve the high ranking. They do--they performed on the field. I am simply arguing that the game has changed such that a more modern hurler is at a disadvantage in being able to make a similar impact over an extended period.
BP: Three players who have been called overrated are the "Tinkers to Evers to Chance" trio. What are your thoughts on each, and do you feel that Deadball Era players are adequately represented in the Hall of Fame?
DL: Frank Chance was clearly the top first baseman of the era. A great player both offensively and defensively, the only knock against his induction is his modest career totals because he played only 1,287 games. Johnny Evers is more problematic--at second base, Evers ranks in the tier well below Eddie Collins and Nap Lajoie, with Miller Huggins, Danny Murphy, Del Pratt, and Larry Doyle, none of whom are in Hall. By sabermetric measures, his fielding was good but not at a level sufficient to move him above his offensive peers. On the other hand, fiery middle infielders were highly valued at the time, especially those who won. Evers's teams won four pennants and two World Series championships; for his season with the Miracle Braves in 1914, he won the NL MVP.
While not without his weaknesses, one can make a decent case for Tinker. His long-service offensive comparables from the era at shortstop include Bobby Wallace, Bill Dahlen, Donie Bush, Kid Elberfeld, Art Fletcher, and Dave Bancroft. Bancroft and Wallace are in the Hall. Offensively, Tinker probably falls towards the lower end of this group. It's on defense where Tinker makes up the difference and becomes a legitimate choice--sabermetrically, on a per-inning basis, he ranks as one of the best shortstops of all time. Care must be taken, however, when drawing conclusions from fielding statistics because the methodology for translating them into a value measure cannot be verified against team totals in a manner similar to offensive statistics.
The table below shows the number of players in the Hall of Fame based on the decade in which they were born. Given that most players who spent a considerable portion of their career in the Deadball Era would have been born in the 1870s, 1880s, and early 1890s, the period appears fairly represented. As an aside, players who starred in the 1920s and 1930s seem overrepresented, a not surprising conclusion for those who have studied HOF induction patterns. Players born in the teens and twenties had a more difficult path to the Hall, as many of their careers were interrupted by World War II. If anything, it's the modern player that appears underrepresented; some born in the 1940s and 1950s still have a shot, but many have been through most of their eligibility and passed over.
Birth Decade HOF Count 1850s 11 1860s 13 1870s 18 1880s 24 1890s 30 1900s 41 1910s 19 1920s 17 1930s 24 1940s 16 1950s 11 1960s 1
BP: How has the importance of fielding evolved?
DL: In the Deadball Era, fielding was a much larger percentage of defense (a combination of both pitching and fielding) than it is today. In one relatively simplistic but telling measure, the percentage of runs that were unearned fell from more than 25 percent of all runs scored to around eight percent today. Of course there is more to evaluating fielding than simply accounting for errors. With fewer strikeouts, walks, and home runs, a larger percentage of plate appearances ended with a ball in play. Observers at the time recognized the importance of fielding, and held star defenders in high regard. Even at the time, many recognized the limitations of fielding percentage as a measure of fielding ability. Several articles in The Sporting News discussed using what we now call Range Factor [(putouts + assists) / games] as a potentially superior measure of fielding prowess.
BP: Which managers had the biggest impact on the game of baseball and how it was played?
DL: Managers were not only responsible for the duties we associate with them today, but many of those performed by the modern general manager as well. He was accountable for many of the team-building functions: finding and signing young talent, as well as making trades. Much like today's college football and basketball coaches, a Deadball Era baseball manager was responsible for both the makeup of his team and the actual performance of his players.
Longtime New York Giant manager, John McGraw, was one of the most dominant baseball personalities of the time. McGraw captured six Deadball Era pennants, and his teams were almost always in the pennant chase. McGraw believed in aggressive, heady play. The key to understanding McGraw's managerial style is to realize that he worked hard to instill two separate (and to a large extent, contradictory) characteristics in his charges. On the one hand, he demanded absolute obedience and loyalty. McGraw felt any challenge to his authority would lead to the team losing its focus and drive, and he could be extremely unpleasant and abusive in trying to instill this single-mindedness. He also, however, demanded a quickness of mind on the ball field. A player was required to learn the McGraw way to play, play it well, and remain obedient. If he did this, McGraw would show him loyalty and pay him well; if not, McGraw could be cruel and vicious. Many of McGraw's players went on to manage, extending his legacy many years.
Philadelphia Athletic manager Connie Mack also won six deadball pennants and had a number of ex-players accept managerial positions. Mack may have had more contacts throughout the baseball world than anyone, and he used them to find his players. A tremendous judge of baseball talent, Mack also looked for intelligence. Mack's firm yet more sympathetic approach to many of his players earned their respect. Both Mack and McGraw enjoyed some excellent seasons into the lively-ball era, but both also suffered more limited success towards the end of their managerial careers.