A Reconstruction of Philadelphia Town Ball

Olympics of Philadelphia, 1837 Constitution

The previous post, Richard Hershberger’s article on the 1863 “New Marlboro Match Baseball Co.”, elicited this comment from reader Jim Roebuck: “One thing I’ve been trying to figure out – and I’ve read a fair amount about it, but I’m still confused – is the difference between town ball and the Massachusetts Game. Topic for another essay?” To this I replied, “The two are substantially different, but modern-day scribes have been calling all bat & ball games other than the New York Game “town ball” for a long time. This has bred confusion indeed, and prompted Richard Hershberger to tackle the subject in the journal Base Ball in the Fall 2007 number. I’ll run his full article, ‘A Reconstruction of Philadelphia Town Ball,’ in this space soon.”

Here ya go.

A Reconstruction of Philadelphia Town Ball

Richard Hershberger 

Modern baseball is descended from the game played in New York City at the middle of the 19th century. This version, however, was not the only one played in North America. The baseball family extended throughout English-speaking North America, in various versions and under different names, both as children’s games and in formal competitive communities of clubs of adults.

The best documented of these other forms is the game played in New England. There arose in the late 1850s extended communities of clubs in both New England and New York, holding conventions and publishing formal rules.

Smaller communities are known to have existed in various cities including Philadelphia, Cincinnati, and Chicago, but none of these published its rules. There is a long tradition of assuming that the game played in these areas was substantially identical to the New England form, but there is little evidence to support this theory. The more conservative belief is that the rules are unrecoverable. A close examination of the evidence reveals, however, that the rules of the game as played in Philadelphia can be reconstructed.

A Brief History of Town Ball in Philadelphia

The American Sunday School Magazine reported in early 1830 that the previous summer a group of 18 adult rope makers met for a game of ball one Sunday afternoon near a Philadelphia orphanage. The matron of the institution remonstrated with them for breaking the Sabbath and invited them into the orphanage to see how the Sabbath was kept there. They heard the orphans sing a hymn, “This day belongs to God alone, He chooses Sunday for his own….” The ballplayers were moved to tears, and sat in perfect silence while hymns were sung, answers from the catechism recited, and verses of scripture repeated. The next Sunday every one of the 18 returned, decently dressed, and witnessed the exercise again. Many returned yet again for a third visit, moved to repent their former ways as Sabbath-breakers. Regardless of the veracity of this tale, it makes clear that the author considered the idea of adults playing ball plausible enough to include it without further explanation. Organized club play appears soon thereafter.

Olympics of Philadelphia clubhouse ca. 1860

In 1831 a group of men in their mid-20s made the ferry ride across the Delaware River to Camden, New Jersey, to play town ball on Saturday afternoons.[1] At the same time a club under the name “Olympic” convened to play town ball on the Fourth of July, and occasionally on other days as well. Following the example of the Saturday group, they began practicing on the same ground on Wednesdays. This led to a match game—among the earliest known, but with the results unrecorded. Following this the two groups merged, practicing two days a week as the Olympic Ball Club. They absorbed two other groups of town ball players over the years, probably in the late 1840s and early 1850s, the latter said to be graduates of Philadelphia’s Central High School. They played in Camden into the late 1850s, when they moved across the river to Pennsylvania.

This summary of what is known of town ball play prior to 1857 comes from two documents: the Olympic Ball Club constitutions of 1838 and 1866, the latter including a brief history of the club.[2] It is obvious that town ball play was not confined to the Olympics, but the evidence has not come down to us. A hint of its existence is given by the Honey Run Club of Germantown.

Germantown had been an independent borough in Philadelphia County, about six miles from the City of Philadelphia and most famous as the site of a Revolutionary War battle. The Act of Consolidation of 1854 unified the City and County of Philadelphia, reducing Germantown’s status to a mere neighborhood. It was still separated from the urban center by farmland and retained its distinct character for many years.

The earliest evidence for town ball play in Germantown is a record of a game played in September 1857. Twenty grown men who had been schoolmates gathered and divided into teams. The event had a nostalgic air to it: The lot “presented somewhat the appearance of other days” and “the old ‘schoolmates’ seemed to enjoy each other’s company as ‘in days of yore.’”[3] Clearly town ball was no longer a novelty at that point. It is less clear, however, if adults playing the game was. The only clear peculiarity was that the game was reported in the press, via a letter by the pseudonymous “Sport.”

The next record of the game’s appearance in Germantown dates to November 1859. The Honey Run Town Ball Club, “consisting of twenty practised members,” challenged the other clubs in town for a match on Thanksgiving Day. The Balsch and the Charter Oak clubs declined. The Honey Run met to prepare for an intraclub game, when two delegates from the Marion Club appeared to accept the challenge—to play for a supper. Both clubs set to practicing at every opportunity, and music was engaged for the day of the match, again reported by “Sport.” The game came off splendidly, the Honey Run winning by two runs in an exciting finish that prompted “Sport” to provide what is by far the most complete extant account of a town ball game (see sidebar). The Honey Run later presented their ball giver, the hero of the game’s climax, with a gold ring at a festive dinner.[4]The Honey Run make one more appearance the following spring, on Easter Monday, playing an intraclub match, this time reported by “Saint,” and some members turned up in the Army of the Potomac playing town ball in 1863.[5] There are no further mentions of the Balsch, Charter Oak, or Marion clubs.

There were, then, at least four organized clubs in Germantown, apparently playing mostly intraclub games and going virtually unnoticed by the press. The reports of the Honey Run’s exploits result from the combination of an enthusiastic correspondent and the rise of a New York sporting press willing to publish such reports. Both games of the Honey Run took place on holidays; one of the groups founding the Olympic club had existed specifically to play on the Fourth of July. The evidence suggests, then, that there was a tradition of holiday play in the Philadelphia area that evolved—perhaps due to rising urbanization—into clubs formed to organize this recreational activity. A modern equivalent is the Philadelphia mummer clubs, which put on an annual New Year’s parade. Their activities entailed preparing for, participating in, and recuperating from this one day.

A different tradition also developed in the late 1850s: competitive club play. This new brand of ball club closely resembled the New York clubs. By 1859 there were at least four such clubs, some fielding first and second teams. The Excelsior club was active at least by 1859,[6] while the Camden club organized in 1857;[7] the Athletic club organized May 31, 1859.[8] They, along with the Olympics, were playing match games at least by 1858, with the Olympics and the Camdens playing three that year.[9] Gone was the old habit of absorbing and internalizing competition. In its place evolved a competitive ballplaying community much like that in New York, but about five years behind New York in its development.

This new brand of Philadelphia town ball was not to last long. On Thanksgiving Day of 1858 the newly formed “Penn Tigers Social Base Ball and Quoit Club” played Pennsylvania’s first New York baseball game.[10] Late in the following season they were joined by the Pennsylvanias, the Nonpareils, and the Continentals.[11] The spring of 1860 saw the fad for the New York game take off. By May there were not fewer than 10 clubs, with more added as the season progressed.[12] The first interclub match game was played June 11 by the Equity and the Winona (formerly the Penn Tigers), the Winonas winning 39–11.[13] In September the Excelsior Club of Brooklyn visited Philadelphia, defeating a picked nine 15–4, bringing Philadelphia into the expanding baseball fraternity.[14]

The competitive town ball clubs joined in the transition to the New York game. The Athletics voted in early 1860 to switch, and they never looked back. There is no record of their ever playing town ball again, and they nearly forgot they had ever done so. Within a year, 1860 was being published as their foundation date.[15] The Excelsiors held out, playing only town ball the season of 1860; but by 1861 they too had adopted the New York game.[16] The Camdens are a cipher; there is no sign of them after 1860. A Camden club appeared several years later playing the New York game, but there is no obvious connection between it and the town ball club.

The Olympics in May 1860 also voted to make the switch. They didn’t abandon town ball entirely and immediately, playing a match game with the Excelsiors and scheduling an intrasquad game as late as 1862.[17] In 1864, New York journalist Henry Chadwick claimed that the Olympics “favor [town ball] almost entirely; and but for a few members would not play Base Ball at all.” Chadwick certainly vastly overstated the case. He had recently been accosted by a member of the club and threatened with violence over his reporting, and his assessment of the Olympics was not dispassionate. Nonetheless, for the assertion to be plausible to its readers would require that the Olympics were, at least to some extent, still playing town ball.[18] Unlike the Athletics, they embraced their early history and the prestige of seniority. (The Philadelphia press was always ready to point out that the Olympics were older than the Knickerbockers.) The Olympics lasted nearly another quarter century, but with no reports of them, or anyone else, playing organized town ball.

“Town Ball” and “Base Ball”

It is necessary to undertake a linguistic digression in order to define what is and is not accomplished by describing Philadelphia town ball. The baseball family of games was, in the mid–19th century, widely played in both North America and in Britain. Not yet standardized, there were innumerable local variants. The games also went by various names but, unlike the variant rules, the number of names was small.

In Britain the oldest name was “base ball,” while the game was known as “feeder” in the London region.[19] Both names died out, with “base ball” being included in a list of archaic words.[20] Their place was taken by “rounders.” In New England the term “round ball” was used in early days, but largely disappeared over the first half of the 19th century. Two names prevailed in North America: The old term “base ball” dominated in New England, New York, and the Great Lakes region, while “town ball” prevailed in Pennsylvania and the Ohio River and upper Mississippi valleys.

With more local variant forms than there were names, it is obvious that name and game did not always represent clear 1:1 relationships. Into the 1860s this was considered unremarkable, and the press published remarks such as “Base Ball at Ingersoll…The game played in Canada differs somewhat from the New York game…” and “Town Ball at Evansville, Ind….the rules and regulations for playing the game of town ball vary a great deal.”[21] The two forms that were standardized in the 1850s were both called “base ball,” so they were distinguished as the “New York game” and the “Massachusetts game.”[22] This was unnecessary with regard to regions using “town ball,” with the New York press using the unmodified “base ball” to refer to the New York game.

Ball, Bat and Bishop, Henderson 1947

The New York game came to dominate all others over the 1860s, so local variants in the old “base ball” regions came to be described as “old fashioned base ball.”[23] In later years, people became uncomfortable applying “base ball” to anything other than the New York game. The name “town ball” was adopted retroactively in regions that had never used the term, including renaming the Massachusetts game. Just as it was assumed that “base ball” could mean only one variant, so it was assumed that “town ball” must also apply to just one form. Even as astute an observer as Robert W. Henderson, the first serious student of the early game, wrote in 1947, “A town ball team was fully organized in Philadelphia in 1833 and it continued to be played in New England until 1860, where it was known as ‘The Massachusetts Game.’ ” A purported description of town ball followed this quotation, but it actually described the Massachusetts game.[24] Fallacy is layered atop fallacy.

So the reconstruction of Philadelphia town ball presented here is no more than that—a reconstruction of Philadelphia town ball. It should not be taken as a reconstruction of any other regional form, including any other variant also called “town ball.” There is no evidence to suggest that “town ball” forms were any more or less similar to one another than they were to variants of “base ball” or “rounders.” This reconstruction applies merely to the Philadelphia region, and the Philadelphia region stands out only in that its town ball is unusually well documented and thus particularly well suited for such a project.

Finally, the terms used in this article are the “baseball family” to refer collectively to the related forms of the game, whether locally called “base ball,” “town ball,” or “rounders”; the “New York game” and the “Massachusetts game” are so called, to avoid ambiguity. The game in and around Philadelphia is called “Philadelphia town ball” or, for brevity and when the context is clear, simply “town ball.”

Sources

The sources used to reconstruct the rules of Philadelphia town ball fall into three categories, in order of decreasing reliability:

Box scores, in particular of three match games: one between the first teams of the Olympic and Excelsior clubs played July 12, 1860, and two between the second teams of the Excelsior and Camden clubs, played July 9 and July 23, 1860.[25] These are elaborate records, including the fielding records and how the players were put out—information not found in modern baseball box scores. They bespeak sophisticated scorekeeping, and are the most objective source of information we have. Box scores of the New York game had not yet been standardized, and ranged from rudimentary records resembling modern box scores, to extended records with fielding and “how put out” records, much like contemporary cricket boxes and comparable to these three town ball boxes. The more elaborate forms were used for important games, implying that the three town ball games were considered significant at the time.

Contemporary game accounts: The account of the Honey Run Club’s game of Thanksgiving Day 1859 is by far the most complete account extant. There are, however, various shorter, fragmentary descriptions that shed light on certain aspects of the game. Narrative accounts are more subjective than box scores, and require more interpretation, but as contemporary texts written by reporters familiar with the game, they are likely to be accurate.

1884 Olympic Club of Philadelphia

Retrospective descriptions: The most important of these is the historical sketch included with the 1866 constitution of the Olympic club, which shares some text with a sketch of the club published in 1861.[26] Reminiscences are inherently suspect, but in 1866 town ball was still a recent memory, with the club retaining members from its town ball days, while in 1861 a description was only barely retrospective. Also notable is a sketch of the Olympic club published in 1884.[27] The Olympics were still a going concern, and the article includes a hint of reference to club records since lost. It is unique for pieces of such a late date in that it does not rely on the 1866 sketch, yet is still consistent with the known facts.

Unifying Assumptions

The previous section emphasized the diversity of the baseball family. Nevertheless, we can still assume a degree of unity amongst the various games.

It is assumed that the competitive Philadelphia town ball clubs played under similar sets of rules. There is no direct evidence for this, but there are no reports of negotiating rules as was sometimes found in other areas. There were conventions among both New York and New England clubs in the 1850s to standardize their rules. There is no hint of a similar convention of Philadelphia clubs in the town ball era. A likelier guess is that the rules of the Olympic Club, as the senior, prestige club, were adopted by the other clubs (in much the same way that the laws of cricket were the club rules of the Marylebone Club in England), or at least that the other clubs adopted field rules with only minor variations. These rules are what this article attempts to reconstruct.

It is also assumed that the noncompetitive clubs such as the Honey Run were playing essentially the same game as the Olympics, although likely in a less formal manner. This is the opposite assumption from that commonly made of the New York game. Modern writers generally acknowledge that some version of the baseball family was long played in New York, but assume that the game the Knickerbockers played as of 1845 was in its essence different from that of earlier generations. The rules of the Knickerbockers have more than their share of peculiarities compared with other members of the baseball family, but there is little direct evidence of the rules under which young New Yorkers were playing in the 1830s. It is not actually known whether the peculiarities of the New York game originated with the Knickerbockers or were inherited by them. That the game of the Olympics was the same as that of the Honey Runs is not provable. Indeed, some of the vocabulary applied to the non-competitive club games is not found in the competitive matches. But unlike pre-Knickerbocker New York baseball, we can compare accounts of competitive and non-competitive Philadelphia town ball and observe that they seem to be similar, and guess from this that the account of the Honey Run’s match can illuminate the Olympic game.

Finally, it is assumed that Philadelphia town ball was a member of the broad baseball family, sharing characteristics of the family. For example, nothing in the accounts of Philadelphia town ball explains what an “inning” is; but since the word is used the same way throughout the baseball family, there is no need to believe that an “inning” in Philadelphia town ball presents any mystery.

The Rules of the Game

The Players: A team consisted of 11 players, unless the clubs agreed to some lower number. A match in 1858 between the Olympics and the Camdens was played with nine on a side.[28] The Germantown games were more variable, as would be expected of the less formal context, with typically 10 or 11 on a side. That 11 was normative, at least among the competitive clubs, is shown by the routine use of “first eleven” and “second eleven” to designate the clubs’ first and second teams. New York clubs used the analogous “first nine” and “second nine,” but the likely source for Philadelphia’s rule was the identical usage—well established by the late 1850s—of first and second elevens of cricket clubs. In November 1859 the Pennsylvania Base Ball Club formed to play the New York game. In their first intraclub game they played 11 on a side, apparently not having carefully studied the New York rules. They realized their mistake, or it was pointed out to them, and by the end of the month they were playing nine on a side.[29]

The only players with assigned positions were the ball giver and the behind, corresponding to the modern pitcher and catcher. It was typical of the baseball family that the other defensive players had no fixed assignments. The New York game abolished the general practice of throwing the ball at the runner and replaced it with tagging the base or runner. This led to the assigning of players to man each base. In other forms there was no need for this, and the players could position themselves as strategy or whim dictated. (A vestige of this can be seen in the modern game, comparing the position of the first basemen with a runner on base versus without.)

The Field: The bases were five stakes arranged in a circle of approximately thirty feet in diameter.  The small size of the playing field is perhaps the most surprising aspect of Philadelphia town ball. The sources on this subject are consistent and clear. The 1861 article on the Olympics noted that some older members retired when the New York game was adopted, since “three hundred and sixty feet, compared with the old town ball circle of eighty feet, was enlarging the sphere of action with a vengeance.” The 1884 account described the circles as “about thirty feet.” A circle 30 feet in diameter has a circumference of about 94 feet. It is likely that the distances were not intended to be precise and, as will be seen, the batter probably did not run the entire circumference anyway.

This is the smallest documented size of any field of the baseball family, with about 19 feet between bases. The Massachusetts game had basepaths of 40–60  feet. In 1828 The Boy’s Own Book by William Clarke described the four-base diamond formation with the bases “placed from twelve to twenty yards asunder.” This is usually interpreted as the length of the basepaths, though the Knickerbocker rules of 1845 defined the size of the diamond as the distance between home and second, and first and third. If the 1828 distance is measured similarly this still results in basepaths more than 25 feet in length. The earliest known rules for baseball, published by Johann Chistoph Friedrich Gutsmuths in 1796, placed the bases 10–15 paces apart, making it the variant closest to Philadelphia town ball.[30]

David Block notes a trend within the baseball family of the field gradually expanding.[31] The Philadelphia town ball field seems to have been a uniquely antique feature retained from the ancestral game.

The use of stakes was one standard option in the baseball family, used in the Massachusetts game and surviving in modern rounders in Britain. The relevant sources agree that stakes were used in Philadelphia. These include the 1884 account, which called them “sticks,” and an 1862 account of the history of the Athletics, which poetically described their switch from town ball to baseball as the adoption of “the bases instead of the stakes.”[32] The account of the Honey Run match mentions flags being placed at the corners. “Corner” is a synonym for base found in some later accounts of early baseball. In 1867 the “Home Run Polka” was published by a Philadelphia publisher, dedicated to the National Base Ball Club of Washington. The Nationals had from their inception played the New York game, but the front page bore a strange illustration of a baseball game—apparently drawn by an artist who had never actually seen the New York game played, and conflating elements of the New York game with older forms. It features stakes misplaced halfway down the basepaths, with small rectangular flags. It is possible the artist recalled these from town ball games. There is no evidence of whether or not they were used by the competitive clubs.

Home Run Polka

The arrangement in a circle is unusual but not without precedent. Early forms of baseball had been flexible about the number of bases. Most later forms standardized this at four bases. The Massachusetts game is conventionally characterized as having five bases, but it actually had four stakes and a designated location for the batter. Placing the batter in the familiar location at home base presents obvious practical problems if home base is a stake. So the batter was moved to the first-base side in those forms of baseball using stakes. The 1884 account of Philadelphia town ball states explicitly that there were five stakes. The only other example of five bases so arranged was described in 1855 in the Manual of British Rural Sports. [Editor's note: Richard wrote this article three years before the unearthing of the New Marlboro rules and diagram, with their five bases. See: http://ourgame.mlblogs.com/2013/05/21/new-marlboro/]

This circular arrangement raises the question of where the batter was placed. Following the pattern of other forms with stakes, he likely was about midway between fifth (or home) and first base. This hypothetical reconstruction also shortens the circumference of the path to run to about 85 feet—close to the 1861 stated distance of 80 feet.

Pitching: There is no direct evidence of where the ball giver stood, but every known form of baseball places him somewhere within the area delineated by the bases. This was so strongly assumed that the Knickerbocker rules of 1845 didn’t bother to mention it. There is no direct evidence concerning whether the ball giver delivered the ball overhand, as in the Massachusetts game, or underhand, as in the New York game, but it was almost certainly underhand.

The Massachusetts game featured a swift delivery, attempting to overpower the batter, while the early New York game featured a soft toss intended merely to put the ball in play. The tiny distance between the ball giver and the batter in Philadelphia town ball seems ill suited for a swift delivery. For the games with full box scores it is readily apparent who the behind was, and on one occasion the behind was singled out for praise for his good fielding. But the identity of the ball giver is usually (and conspicuously) omitted in these accounts. Wyn Stokes, the Honey Run’s ball giver, was the hero of their game against the Marions, but for making the game-ending defensive play, not for his pitching.

Finally, there is once again the negative evidence of silence regarding a possible change in pitching style when the New York game was adopted. This difference was a topic of comment in comparisons between the New York and Massachusetts games, so one would expect the subject to have arisen had there in fact been substantial changes.

This all suggests that the delivery was a slow, underhanded form similar to that of modern slow-pitch softball.

The ball and bat: The 1884 article states that the ball was “much lighter and softer than the ball of the present time.” It also states that the players had to make their own balls and bats, and there indeed is no evidence for commercial manufacture of town ball implements. Under the circumstances, it is likely that the balls varied widely, and were probably quite dead.

Various forms of bat were found in the baseball family. Two-handed round bats were used in both the New York and Massachusetts games, but other forms included one-handed bats (used in modern rounders) and flat bats (like those used in cricket) of various sizes, often described as “paddles.” Once again there is an absence of discussion on the subject, suggesting that the town ball bats were two-handed round bats like those used in the New York and Massachusetts games.

Ending the inning: The inning was ended when every player on the side had been put out. This is one of two common versions in the baseball family, the other being the inning ending when one player was put out. The New York game’s feature of ending the inning after three outs was unique.

Getting out: The three 1860 box scores feature “How Put Out” headings for each out, divided into five categories: Fly, Bound, Behind, No Balls, and Stakes. There also is a section recording each player’s fielding record, listing Fly, Bound, and Behind. Modern baseball scoring rules require that every out be credited to a fielder, but Philadelphia town ball felt no such obligation, suggesting that No Balls and Stakes were not considered fielding accomplishments. Table I below lists the percentage of outs made in each category, while Tables 2 and 3 split these: the July ¡2 game between the first elevens of the Olympics and Excelsiors in Table 2, and the July 9 and July 23 games between the second elevens of the Excelsiors and the Camdens in Table 3, with the differences attributable to the varying skill levels of the first and second elevens.

How Outs Were Made

“Fly” outs are exactly as they are in modern baseball. The fly out is a universal feature of the baseball family, and even extends beyond it to cricket.

“Bound” outs are balls caught on the first bounce. Philadelphia town ball shared this feature with the contemporary New York game.

“Behind” outs are more mysterious. The breakdowns of outs by fielder lead to the unsurprising conclusion that these outs are credited to the behind (i.e. the catcher). In the July 12 game, all but two of the 70 behind outs are credited to the two players identified in the account as the behinds (the other two presumably made during a defensive switch), and the two behinds each made over twice as many outs a the next-most productive fielder. But what, exactly, was a “behind”? The answer may lie in the apparent fact that the behinds are never credited with other sorts of outs (except when the box score indicates a defensive switch). One would expect the behind to be in a position to catch pop flies and, on the bound, balls tipped backwards into the ground. Both of these were common ways for New York–game catchers to make outs. So the “behind” category may have been created for statistical purposes, distinct from fly and bound outs because the behind had so many more opportunities. A reverse analogy of sorts might be found in the way modern baseball scoring has the distinct categories of passed balls and wild pitches, rather than including them simply as errors.

Fly, bound, and behind outs account for more than 90 percent of the recorded outs. The remaining categories played relatively minor roles.

“No Ball” is a somewhat confusing term. It is not found elsewhere in the baseball family, but it occurs in cricket. A cricket “no ball” is a ball bowled beyond the reach of the batsman, and the fielding team is penalized. The Philadelphia town ball “no ball” is detrimental to the batting team, so it clearly is not the cricket “no ball.” More likely, these outs correspond to the modern strikeout. This was an old feature of the baseball family, going back to the 18th century.[32] If the batter swung at and missed three pitches, he was ruled out if the third pitch was caught, but the ball was considered to be in play if the behind failed to catch it. This is the origin of the modern dropped–third strike rule, and the Massachusetts game had a similar rule. With soft deliveries, “no balls” were not a major feature, accounting for about 8 percent of all outs, and only 5 percent in the first elevens’ games.

“Stakes” are probably, if only through the process of elimination, fielded balls thrown at the runner, striking him between bases. This was a widespread feature of the baseball family, often called “soaking” or “plugging” the runner. Abolishing the practice was one of the major innovations of the New York game. Stakes were quite rare in Philadelphia town ball for reasons that will be discussed below, accounting for less than 2 percent of all outs. But as the account of the climax of the Honey Run–Marion match shows, they could be dramatic.

Running the bases: Running the bases on balls in play is a universal feature of the baseball family. Philadelphia town ball has the variant that the batter lacked the option of stopping at a base—every at bat resulted in a home run or an out. The 1884 account is explicit about this: “The striker was compelled to make a complete circuit upon each hit in order to score.” Every other known member of the baseball family allows station-to-station advancement through the bases. This and the small field size are the two striking peculiarities of Philadelphia town ball, and clearly are connected with one another. The entire circuit was shorter than the modern distance from home to first base, and advancing station to station would be a trivial achievement.

Several accounts of Germantown games mention “grannies.” One of them reveals that this is a score, distinguishing between “regular circuits” and “grannies,” with over 10 times as many regular circuits as grannies.[34] The captain of the Honey Run club in their match against the Marions was described as a “granny runner.” A possibility is that grannies were scores which required the runner to dodge an attempted stake, while regular circuits were made on unfielded balls. No account of the competitive club matches mentions grannies or makes any distinction between different types of runs.

The requirement to make a complete circuit on each hit raises the related question of whether the bases served the role of safe havens. This is a general feature—so much so that it sometimes is considered a defining characteristic of the baseball/cricket family. But if the runner must continue running, the safe-haven status seems moot. The Honey Run–Marion account suggests that the bases did retain this function, if only vestigially. In the climactic play of the game, Righter of the Marions hit the ball and before he had reached the third corner (i.e. halfway around the bases), Wyn Stokes of the Honey Runs had the ball in his hands. “This was a critical time…every player was nervous with excitement. The marksman stood still; Righter afraid to move. Wyn’s arm drew back, and with terrific force, catching ‘Marion’ (just making a fine dodge) about three inches above the ancle [sic] bone.” Was the immobility of Righter and Stokes a tactical decision? Should we chalk it up to nerves? Or was it mere dramatic license by the chronicler? Taking it at face value, one possible interpretation is that Stokes didn’t throw immediately because Righter was touching a base, then Righter made a break for the next base, unsuccessfully attempting to dodge the throw. On the other hand, the scene is very cinematic, and possibly fictional. It seems somewhat likelier that the bases were vestigially safe havens, but by then the runner could not linger and allow the fielder with the ball to approach him.

The complete-circuit requirement removes one of the difficulties inherent in all-out innings: what to do when the batter is stranded on a base. In cricket there must always be two men on offense, so while there are 11 men on a side the inning is over after 10 outs. Many versions of the baseball family have special rules for the last man, typically allowing a home run to cancel the previous outs, restarting his side’s inning. Philadelphia town ball had no need for any such rule, there being no mechanism for stranding a runner.

This also raises a question about the batting order. Box scores clearly show a set batting order, but there is no indication whether a successful batter returns immediately to bat, as in cricket, or if the next man in the lineup takes his place, as in the New York game, cycling through the shrinking roster of batters not yet put out. The latter is more consistent with other members of the baseball family, but the former seems a better fit to this version. That said, there is no direct evidence, since no account combines a batting order with the name of the final batter.

A final possible nicety is that the runner was not required to touch the bases as he went by. In 1862 a reporter from the New York Clipper, probably Henry Chadwick, accompanied a group of Brooklyn players to Philadelphia. His report included advice on fine points such as the need to put down chalk lines to delineate the foul lines, and that the Philadelphia players needed to learn to touch the bases.[35] One reason for this requirement is to prevent the runner from cutting corners; but the town ball stakes would clearly define the circle outside of which the player must run. The Clipper’s advice may indicate a vestige of town ball play.

The Umpire: The office of umpire was much less important than in the New York game. The 1838 Olympic constitution assigned this duty to the scorekeeper, who was a club officer, but this obviously would have been inadequate for matches between clubs. The Honey Run–Marion match had two umpires and a referee.[36] Early New York matches followed the same pattern, with an umpire from each club and a neutral third party should the umpires be unable to reach agreement. It soon became apparent that the club appointees were superfluous and a single neutral umpire became standard. Competitive Philadelphia town ball clubs apparently followed the same progression, as the one mention of the office, in one of the Excelsior–Camden matches, named a single man in the role.[37]

There are no descriptions of complaints about the umpire, nor any of the admonishments against this behavior so common in the New York game. The Philadelphia players were not more virtuous: Such descriptions and admonishments appear soon after the New York game was established there. The New York–game umpire was (and is) called upon to make many close judgments, even before the advent of called balls and strikes: Did the ball land foul or fair? Did the ball arrive at the bag first, or the runner? The Philadelphia town ball umpire’s task was less challenging, with the occasional decision on whether a ball was cleanly caught or the rare staking of a runner. This lesser responsibility brought with it fewer complaints.

Ending the game: How to know when the game is over is normally a straightforward question, but there is no obvious answer in this case. There are two common solutions: playing to a fixed score or playing a fixed number of innings. Both are found in the baseball family. The early New York games played in Massachusetts played to a fixed score: 100 and 21 runs, respectively. The New York game switched to the modern nine innings in 1857. Town ball in Cincinnati was played to four innings.[38] Cricket was played by innings: one or two, depending on what the clubs arranged. So either scheme was freely available for Philadelphia town ball, but neither seems to have been applied. Competitive matches came in variously at 11, 12, and 19 innings. The Honey Run–Marion match lasted a mere two innings. The line scores, where available, make clear that extra innings to break a tie were not at issue. Had games been played to a fixed score, however, one would expect it to have been a round number—but this is clearly not the case. Recorded scores include 119–81, 85–75, 80–42, 87–71, and 71–66.[39]

The remaining possibility is that they played for a predetermined period of time. Several accounts of both competitive match games and of Germantown games mention the time of play, consistently running about four hours. The Olympic–Excelsior match of July 12, 1860, ended at 6:30. This is too early to be forced by darkness (even before the advent of daylight savings time), but it is a splendid time to stop for supper. Playing to a fixed ending time is uncharacteristic of the baseball family, but it was the de facto rule even in modern baseball before the advent of lights, and it was not uncommon for teams to agree on an ending time in the days when teams had trains to catch. It is likely that this was the de jure method of ending Philadelphia town ball games.

The Course of Play

The fundamental skills of Philadelphia town ball were identical to those of the early New York game: throw the ball, catch the ball, hit the ball. When the newly formed Mercantile Club lost to the Athletic Club, the Mercantiles ascribed their loss to the Athletics’ experience playing town ball.[40] Based on the evidence at hand, the players’ argument seems entirely plausible. The Olympics were clearly the champion town ball club of Philadelphia, and—following their adoption of the New York game—their title of champion was acknowledged to carry over to the new game.[41] Their play bore this out, and they successfully defended their title for several years.

In strategy, Philadelphia town ball falls short of its relatives. Indeed, it is difficult to see where there is any strategy at all, if we take “strategy” to mean adjusting one’s style of play according to the game situation. In Philadelphia town ball, a “plate appearance” can produce only two possible results: a score (and the right to a subsequent attempt) or an out. There is no possible benefit to be accrued by sacrificing an out, and no partial benefits of getting on base or additional benefits of driving in multiple runs. So there is no reason to bat differently in different strategic situations, and while the fielding team might make adjustments for stronger or weaker batters, this should technically be considered “tactics” rather than “strategy.”

The vast majority of outs were made on fly balls, either caught on the fly or the bound. Unfortunately, there is no record of how most of the runs were made: on uncaught fly balls or on ground balls. The first run of the Honey Run–Marion match was made on a ground ball, but we don’t know if that was common or rare. Given how rare stakes were, it seems that hitting ground balls would be the surest approach. Or perhaps stakes were rare because the fielders played close in, making ground balls dangerous, and batters preferred to hit fly balls. It is a good bet that the ball was dead by modern standards; but one of the vast uncertainties is how dead it was, and, by extension, how far the fielders had to spread out.

The How Put Out box reveals the difference in skill level between the first and the second elevens. The first elevens were twice as likely to catch a ball on the fly, while the second favored catching it on the bound. Journalists discussed the relative merits of the two in the New York game. Much of this discussion was ideological, with fly catches judged more manly, but part of it was pragmatic: On uneven ground the fielder could not count on a true bounce, so a fly catch was, when possible, the more reliable play. Whether for practical or ideological reasons, more skillful players preferred to catch balls on the fly.

Similarly, the second teams had more trouble putting the bat on the ball, being twice as likely to be put out on a “no ball.” But even with the second elevens these represented only 10 percent of all outs. This is low by modern standards, but comparable to strikeouts in contemporary New York games, as can be seen in the extended box score for a New York game between the Olympic and Hamilton clubs, with the clubs combining for five strikeouts.[42]

The Honey Run–Marion match took approximately the same time as the competitive matches, and the final score was similar; yet it lasted only two innings, compared with the 11 or so innings of the competitive matches. There are two aspects that require explanation: why the scoring per inning was so much higher in the Honey Run–Marion match, and why the innings took so much longer to complete.

The likely explanation for the high scoring per inning is that the fundamental skill of batting (in a slow-pitch era) was easier than the fundamental skill of fielding (in an era long before fielder’s gloves). Henry Chadwick long held, in the New York game, that a low score—indicative of skillful fielding—was the true measure of a well played game. This is often regarded today as quaint ideology, and Chadwick undoubtedly held on to the idea long past the time when the game was more about pitching and hitting than fielding. But in the earlier era, the capability of amateur players reliably to catch a ball could not be assumed.

As to the length of an inning, the competitive matches maintained a furious rate of activity. The Olympic–Excelsior game had 240 outs in 11 innings (two short of the expected 242 because the Olympics played the first inning shorthanded) and 158 runs, for a total of 398 game events (defining a “game event” as either a run or an out) in the recorded four and a quarter hours, or a game event approximately every 38 seconds. This is not counting unhit balls and not taking into account the time taken to exchange places between half innings. The Honey Runs and the Marions, being more social than competitive, may simply have played the game at a leisurely pace.

The sum of the evidence strongly suggests that the competitive matches were played more skillfully and more aggressively and, simply put, more happened.

Conclusion

The Massachusetts game is frequently taken as being representative of the baseball family as a whole, which is assumed to be largely uniform. The New York game in turn is assumed to be an outgrowth of this, with certain innovations.