“People ask me what I do in winter when there’s no baseball. I’ll tell you what I do. I stare out the window and wait for spring.”

Rogers Hornsby

The season’s end is always a bittersweet time. While I’m not usually one to wax poetic on the mysteries of baseball and look for metaphors in what is, after all just a game, I do find the rhythm that the game reinforces a comfort to my mind. After a long season that begins as the weather heats up, extends through the dog days of summer, and finally culminates in the crescendo of the playoffs, it does seem appropriate that the season should end as the days grow shorter and we experience (at least at my house during Game Four) 16 inches of new snow.

First off, to keep us all from taking Hornsby’s advice, this week we’ll start our offseason musings by taking a closer look at rainouts, the long season, and a modest solution.

Rain, Rain, Go Away…

As you’re all well aware of by now, this postseason will likely be remembered for the Cardinals somewhat improbable run in becoming the team with the fewest wins (83) and lowest winning percentage (.516) to ever the win the World Series. The previous bottom ten World Series winners are listed in the table below.

Year    Team           W    L    Pct
1987    Twins          85   77   .525
2000    Yankees        87   74   .540
1974    A's            90   72   .556
1980    Phillies       91   71   .562
2003    Marlins        91   71   .562
1985    Royals         91   71   .562
1990    Reds           91   71   .562
1959    Dodgers        88   68   .564
2001    Diamondbacks   92   70   .568
1996    Yankees        92   70   .568

You’ll notice that five of the top eleven are from the Wild Card era that began in 1995. This simply reinforces the fact that in baseball, where the outcome of any single contest is based to a greater degree on luck and the performances of individual players, anything can and will (eventually) happen in a series of short series. That of course leads to a discussion of changing the playoff format to hamper Wild Card teams, but that’s a topic for another writer to tackle.

Beyond that observation, the 2006 postseason will also be known as one of the soggiest on record. All told, four games were postponed because of rain including Games One and Five in the Cardinals/Mets NLCS, Game Two of the AL division series between the Tigers and Yankees, and of course Game Four of the World Series. That number matches the previous total of postseason rainouts since the Wild Card format was introduced in 1995.

As a result, on a more historic level, this year will be remembered alongside 1975, when a memorable Game Six in Boston between the Red Sox and Reds was postponed for three days because of rain; postseason play in 1962, when there were three straight postponements as the Series shifted from New York to San Francisco (there was also an additional rainout in the Series); and finally 1911, when Game Four was postponed by six straight days of rain in Philadelphia as the A’s battled the New York Giants.

All of this got many folks to thinking about the topic of postponements in general, and one such intrepid soul is fellow SABR member Paul Rivard, who shared some research he’s done on the question with SABR’s listserv last week.

To summarize, Paul has access to both original and as-played schedules in a spreadsheet. He then uses the online service ProQuest to sift through the New York Times baseball section for each of the 170 or so days of the season and identify which games were actually postponed. Easier said than done, I’m sure. By indicating the postponements in his Excel files he can then run basic queries to calculate the percentage of games that were postponed. As a SABR member Paul has, for the time being, access to ProQuest through his membership. Unfortunately, that organizational access expires at the end of the year, so he’s busily working his way through as many seasons as possible.

As Paul mentioned to me in an e-mail, this data is preliminary, and there are bound to be small revisions in the numbers once he has more time to comb through the data in detail. For example, complete results will also include make-up games that were once again postponed. Of course, these shouldn’t significantly change the numbers, but keep in mind that this is a bit of a work in progress. In addition, this data will also be combined with that from other SABR members and should eventually be made available on Retrosheet as part of the Game Schedules effort.

With those caveats in mind, beginning with 1926 here are the seasons that Paul has collected, along with a graph showing the trend.

    Year    Games   Postponed     Pct
    1926     1232         137   11.1%
    1927     1232         164   13.3%
    1928     1232         156   12.7%
    1929     1232         143   11.6%
    1930     1232         123   10.0%
    1931     1232         154   12.5%
    1932     1232         128   10.4%
    1933     1232         161   13.1%
    1934     1232         137   11.1%
    1935     1232         178   14.4%
    1936     1232         111    9.0%
    1937     1232         133   10.8%
    1938     1232         154   12.5%
    1939     1232         125   10.1%
    1940     1232         162   13.1%
    1941     1232          99    8.0%
    1942     1232         128   10.4%
    1943     1232         106    8.6%
    1944     1232          92    7.5%
    1945     1232         151   12.3%
    1946     1232         121    9.8%
    1947     1232         146   11.9%
    1948     1232         115    9.3%
    1949     1232          71    5.8%
    1950     1232         114    9.3%
    1951     1232          89    7.2%
    1952     1232         123   10.0%
    1953     1232          98    8.0%
    1954     1232          72    5.8%
    1955     1232          77    6.3%
    1957     1232          52    4.2%
    1958     1232          72    5.8%
    1964     1620          69    4.3%
    1969     1944          80    4.1%
    1975     1944          77    4.0%
    1982     2106          57    2.7%
    2004     2430          48    2.0%
image 1

Although there are some chunks of time missing as marked on the graph using the red lines (for example, 1976-1981 and 1983-2003), the graph clearly shows the overall trend. Postponements (not all of them rainouts) have steadily decreased from the 1920s onward, although showing occasional spikes indicating wet years like 1940, 1945, 1947, and 1952. Taking it by decade, the approximate averages are:

Decade  Pct
1920s   12%
1930s   11%
1940s   10%
1950s    7%
1960s    4%
1970s    4%
1980s    3%
1990s   N/A
2000s    2%

The trend is unmistakable, showing a steadily decreasing number of postponements with each decade that seems to have picked up speed in the 1940s and 1950s before slowing in more recent years. The reasons for the trend, on the other hand, are less clear. Several SABR members chimed in with their theories, the most plausible of which were:

  • Better Groundskeeping. Over time, the playability of grass fields in adverse weather conditions has obviously improved. Although tarps were reportedly introduced as early as the nineteenth century, one thinks that perhaps they weren’t used as religiously as they are today. In fact, in the early days it’s been said that the grounds crew might just as well use kerosene or gasoline to light the infield on fire in an effort to dry it, although you shouldn’t use that technique today. More important, though, are the contributions of drying agents such as “Quick Dry” and others, not to mention higher quality turf and better drainage schemes built into the fields themselves, as evidenced by the new system installed at Fenway Park following the 2004 season.
  • The Introduction of Lights. Less obviously, the reduction in postponements may be related to the advent of lighted ballparks in the 1930s, with the first night game being played on May 24th 1935 in Cincinnati’s Crosley Field, and 15 of the 16 teams playing their first night game by 1948. The idea, a logical one, is that the availability of lights would enable teams to wait longer to start an afternoon game without fear of it being suspended by darkness. Interestingly, however, one SABR member commented that until the early 1950s teams were prohibited from turning on lights during a game, with the result that it was not uncommon in a twilight doubleheader (which were much more frequent during that time) for the first game to be suspended by darkness while the second was played with the lights. Even so, the availability of lights starting in the 1950s may help explain the larger drop at that time.
  • Geography. A third contributor is certainly the westward expansion of the Major Leagues, beginning in 1958 with the Dodgers and Giants relocating to California. As more teams cropped up on the West Coast, a greater percentage of the games were played in environments where the weather is less rainy and postponements therefore less likely driving down the overall percentage.
  • Stadium Construction. The 1960s saw the introduction of both domed stadia and artificial turf, starting with the opening of the Astrodome in 1966. The 1970s saw an increasing number of stadia with artificial surfaces that are less prone to rainouts, and from the 1980s forward we’ve seen an increasing number of domed ballparks (Minnesota) and retractable-roof parks (Toronto, Arizona, Seattle, Houston, and Milwaukee), where postponements are a thing of the past.

It’s likely that all of these factors have played a role, with the happy result that 98% of the time when you pack your family in the car, pay for parking, and shell out oodles of money for hot dogs and Dippin’ Dots, you’ll see baseball played.

Let’s Play Two!

As a companion to the topic of this soggy postseason, we’ve heard plenty of commentary along the lines of, “Well, that’s what you get when you play baseball in late fall in the Midwest and on the East Coast–lots of bad weather.” Yet in my eyes it’s difficult to construe the first really weather-impacted postseason in 30 years as a problem.

Before Game Four of this year, just 3.7% of World Series games had been postponed due to weather. That said, these rants were often closely followed by “solutions,” including moving the World Series to a neutral warm-weather site, much like the Super Bowl. That’s all that baseball needs–a ballpark full of disinterested fans during the most important moments of the season.

Along those lines, however, I believe there is merit in at least one suggestion that’s been floated: bring back that venerable baseball tradition, the doubleheader.

If you get in your wayback machine, you might remember when doubleheaders or “twin bills” were relatively common fare and actually placed on the schedule instead of being reserved exclusively for the makeup of a postponed game. Sunday doubleheaders were the rule, as were two games on holidays. Of course you’d have to go back to the early 1990s for the last time that they were a part of the schedule. Since that time, the twin forces of the owners not wishing to “give away” games for “free” and (one assumes) the pressure put on by the Player’s Association in not wanting to subject their players to longer work hours have conspired to make the regularly-scheduled doubleheader extinct.

What we’ve lost is all the attendant lore of the doubleheader: for example, Nate Colbert‘s five home runs and 13 RBI in a doublheader in 1972, commemorated in one of the first baseball cards I ever owned because his achievement duplicated the feat of Stan Musial in a 1954 doubleheader, with the added bonus that Colbert, then eight years old, had seen Musial do it from the stands at Sportsman Park. Likewise, there are all kinds of associated oddities that are excellent trivia fodder: Wilbur Wood starting and losing both ends of a twin bill against the Yankees in 1973, or Max Flack getting traded by Cubs to the Cardinals for Cliff Heathcote between games of a double dip in 1922, where the players switched uniforms and played the nightcap for the other side.

To give you a feel for just how steadily the doubleheader has been disappearing from the game, consider the following chart that shows the percentage of games played in doubleheaders (both ends of the twin bill are counted) in the Majors Leagues from 1920 through the 2005 season.

image 1

As you can tell, doubleheaders actually were less popular in the early 1920s, accounting for around 23% of all games. That changed quickly, and by the mid 1930s–during the cash-strapped Great Depression–we were up around 35%. One can imagine that World War II was the driving factor in scheduling that caused the percentage to peak at 49.5% in 1945, before immediately dropping back to the mid-30s, and then gradually decreasing into the mid-20s throughout the 1950s and 1960s. That decline continued, and 1979 was the last year in which as many as 10% of the games were played as a part of a doubleheader (11.1%). The numbers subsequently dwindled to the point where there were just 1.5% (or 36 games in 18 doubleheaders) out of 2,431 played in 2005. Of course, all of those came as the result of postponements.

As most readers might have guessed, the modest suggestion for avoiding the late fall weather in colder climates is simply not to play during the final weeks of October and have each team schedule one Sunday doubleheader each month (or if the financial is deemed too great, make it a day-night doubleheader on another night with separate admissions). Assuming the complexities of the schedule could indeed accommodate it, those additional six days freed up on the schedule would allow the season to end a week earlier, and would mean that the World Series would be completed around October 20th.

From a historical perspective, this suggestion would result in the same percentage of games played in twin bills as we saw in the early 1980s. As a bonus, we’d all get to once again take part in a great baseball tradition.