I have been a long time lurker and occasional commenter at LL; this is my first attempt at a FanPost. Any feedback and suggestions are greatly appreciated.
It seems that for as long as I can remember, every time we play in Cleveland in the first half of the season, it is pouring rain. I am certain that this is not the case, but I always feel that early season games east of the Rockies frequently are preceded by checking my Twitter Feed to see an unwelcome tweet such as this from Shannon Drayer:
TARP! And brrrrrrrrrrrrrrr! pic.twitter.com/ZRTsYkgpWT— Shannon Drayer (@shannondrayer) April 29, 2014
Juxtaposed almost invariably against this:
Anecdotal evidence even seems to suggest to me that the baseball season tends to be far nicer in Seattle than in many East Coast baseball cities. My recollection as an adolescent in the Pacific Northwest was that it rained basically until July 5th; the joy of setting off street fireworks at the neighborhood 4th of July BBQ was in constant threat of being canceled. After the morning of July 5th rolled around, summer became a near constant stream of 75-85 degree days with not a cloud in sight. I spent some time in Washington D.C. the last two summers and was astounded at the sheer number of days which featured dramatic heavy downpours and evening-ruining thunderstorms.
I have heard grumblings that one of Safeco Field's hallmark features – its umbrella-style roof – is a relatively wasted investment because of the pleasant summer weather in Seattle, but I finally decided to find the numbers to really see how well MLB cities stack up in the rain department.
The Pacific Northwest is a difficult part of the country to adequately classify under climate classification schemes, but Seattle proper is typically classified as an Oceanic climate with dry (Csb) summers which are dominated by a stable high pressure system that dries the air out and keeps humidity relatively low, leading to a very pleasant summer with plenty of sun and minimal rain. Indeed, according the University of Washington's Climate Impacts Group, the Pacific Northwest receives two-thirds of its rain between October and March – conveniently, these are the months that baseball takes a hiatus from the City of Seattle with four glorious exceptions. This is visually represented well in the following chart, also from the University of Washington.
The rest of the country tends to be more prone to heavier downpours. In order to take a deeper look, I took to Wikipedia's collection of climate statistics and built a database of different precipitation levels for each of the cities that house a Major League Baseball team; I used Fort Worth as a proxy for Arlington due to missing data and kept Chicago, New York and Los Angeles as single points for each stadium due to difficulty finding exact rainfall average for each individual ballpark. The first challenge was to examine which cities featured the most rain throughout the baseball months (April - September).
via i.imgur.com (Click to enlarge – does not open in new window).
The graph above shows the total average number of days where precipitation exceeded 0.01 inches (including snow). I ordered the cities from most to least precipitation and greyed out the bars for cities which have a stadium with some sort of roof. The results are not particularly surprising, Seattle is relatively middle of the pack with about 53 average days of rain during the baseball season, falling at twenty-first in the MLB. Safeco Field's roof looks to be a somewhat decent investment compared to the average stadium which sees 50.5 rainy days; however, the dramatic drop off between Arlington and Phoenix skews this average a bit. Most major league teams actually hover closer to 63 days of rain, a full ten days more than Seattle sees. In the approximately 185 days that the Major League Baseball season plays out, most teams have about a 34% chance of scheduling a home game on a day with rain. Total rainy days varies as the season goes on, as shown below.
Looking at the graph of frequency of rainy days across the season, many eastern teams have higher frequencies of rain in the early part of the season (Pittsburg, Cleveland, Cincinnati, New York), while others increase as the season goes on (Miami, Tampa Bay, Houston). This provides some evidence in favor of my gut feeling that Cleveland always seems to have more rainouts early in the season, but frequency of rain is only one component of the rainout equation.
Rainouts do not simply occur because of the presence of rain, the amount of rain is also critical to a rainout. The next graph examines the total rainfall for MLB Cities during the baseball season.
The picture is beginning to become more clear. Seattle – with an average of 9.3 total inches – falls squarely in the bottom 10 cities for most total rain during the season, well below the MLB average of 18.66 inches. Southern parks such as the Florida stadiums, and northeastern and Midwestern stadiums tend to hold on the higher end of the spectrum.
Finally, I wanted to examine the average rainfall during days that it did happen to rain. Unfortunately, the data I had did not allow me to see the exact inches per day of rain so I had to make a rough approximation by dividing each month's total rain by the number of days it rained. I then broke this number up by half of season to get a more robust picture.
Examining the total rain per day of rain puts Settle in a solid 28th place in all of Major League Baseball, behind Phoenix. Furthermore, three of the seven roofed ballparks fall in the top three (while Phoenix is relatively irrelevant to this point as the roof on Chase Field is designed more for cooling purposes than rain protection).
So what do we learn from this? First off, we learn that Safeco Field's roof, while a staple of the ballpark, is one that is relatively extraneous when compared to the needs of most ballparks in the MLB. Rather than being an essential feature to maintain Seattle's ability to continue playing baseball through the season, the retractable roof is a more cosmetic feature which would be better suited from a functional standpoint in an eastern ballpark such as the $1.6 Billion New Yankee Stadium. Furthermore, it appears that many stadiums are more prone to persistent rain in the first half of the season. It may make sense to have these teams play a more road-heavy schedule in the first part of the season in stadiums which are less rain-prone in April and May to prevent rainouts. Based purely on my own conjecture, it makes sense that concentrating home games in poor-weather ballparks during the mid-summer months would also provide a financial boost when baseball is no longer competing with other sporting events (NBA Playoffs, Stanley Cup Playoffs) and becomes the only big-four show in town. Furthermore, the better weather should reasonably drive higher attendance and revenues, proving to be a better allocation of a team's home scheduled games.
There were a number of other analyses I wanted to run on this data, particularly to examine the total number of postponed and rescheduled games due to inclement weather, and the frequency of roof usage, but this data is not available to me. In the future it would be particularly interesting to see if we can estimate the financial or pitching rotation impacts of a postponed game and examine whether league scheduling would therefore play an increased role in both the profits of a franchise or the longevity of its starting pitching staff.