Weather can help make or break the fan experience at a baseball stadium. Kansas City has some of the most variable weather in the country, making it particularly difficult to plan ahead for comfortable conditions at Kauffman. Moreover, weather is a hot topic in baseball circles regarding whether the league should do anything about the potential for cold spring weather to influence games and scheduling. With this in mind, I decided to explore the range of potential weather conditions for each of the Royals’ 81 home games in the 2018 season. To do this, I recruited my wife, whose data science & programming skills are superior to mine. We often work with weather data when developing planting or other management plans on our central Missouri farm, so working with data for the Kansas City area wasn’t a huge stretch.
For this project, we used the National Weather Service hourly normals dataset for 1981–2010 (as recorded at the Kansas City International Airport) and isolated the 3-hour window representing each game, using RStudio for all analysis and plotting. For a more detailed explanation of data & methods, see the Data Notes at the end of this story, but for now keep in mind that the values and ranges shown do NOT represent the full potential for weather extremes and do NOT represent an actual forecast for a given date in 2018.
What constitutes “comfortable” is part science, part psychology. According to Scientific American, the human body functions best when the air temperature is around 70°F, allowing excess body heat to dissipate without unnecessary cooling. However, we all know that individuals vary in their physical and mental tolerance for heat and cold: residents of Mississippi or Montana may laugh at a Missourian’s notion of a hot summer or cold winter day, and locals anywhere can disagree on what’s “right”. Thus, we didn’t try to isolate a given comfort range, just presented all the data so readers could see where their own comfort range fits into the season. It’s also worth recalling that temperature affects gameplay itself; The Atlantic points out that not only do hotter conditions allow balls to travel farther, they can actually lead to hotter tempers and more hit batters.
Spring is particularly volatile, with highs extending well into the summer normal range and lows reaching “INDOOR STADIUM NOW!” levels. In a cold spring like 2018 that has forced numerous game cancellations, it’s worth remembering that spring in Kansas City can also mean lovely summer-like conditions. It’s also clear how much effect the setting sun’s position has on game temperature, as there’s often a bigger gap between 6 and 7 p.m. games than between afternoon (1 or 3 p.m.) and 6 p.m. games.
Of course, temperature is only part of the comfort story; there’s a significant difference between an overcast 85° day and a blazingly clear 85° day. The graph below of normal cloud-cover percentages during game times shows that from spring through summer, cloudiness in Kansas City tends to vary inversely with temperature in a way ill-suited for human comfort. Clouds are most prevalent when it’s already cold (and solar radiation would be desirable) but are far less abundant or likely on those blazing summer days when they would be most welcome. This pattern changes in fall, during which temperatures become moderate or even cold, but cloudiness remains relatively low, allowing for some welcome solar warmth.
Like temperature, wind can affect both fan comfort and baseball behavior. Average wind speed in the Kansas City area is consistently higher in afternoon than evening, dropping off quickly as the sun sets. Although this pattern should be familiar to any resident who spends much time outdoors, we were impressed by the dramatic different between the tightly clustered day game wind speeds and the rapidly declining evening game wind speeds. On average, wind behaves in opposition to fan comfort, tending to be highest in spring and fall when it can add to the wind-chill effect but declining during the summer when air movement would actually be beneficial. On the other hand, what breezes do blow in summer tend to be stronger during the otherwise-hottest afternoon period.
Planning for the season
So with all these data, when’s the best time to head for Kauffman Stadium this year? If you’re a cool-weather person like me, it’s definitely not June through August, but many people love the feeling of a cold beer at the ballpark on a simmering evening as the sun sets and the wind dies sown. It’s worth noting that the current three-month temperature outlook from the National Weather Service has the Kansas City area at a ~40% chance of above-normal temperatures in that period, though all such long-range forecasts should be taken with a popcorn bag’s ration of salt.
Background: The NWS hourly normals dataset is pre-processed (not raw) and presents data values (mostly averages) for each hour for every day of the year. “Normal” is misleading in this sense because, in this region, the true “normal” is a high degree of variability around an idealized but rarely achieved average. This means that extremes are not well-represented in this dataset because it doesn’t present specific record values and excludes values that occurred before 1981 or after 2010. However, using this dataset does allow us to present a reasonable potential range for specified weather conditions within a given time range without excessive analysis of raw data.
Temperature: In this dataset, each hour of every day is represented by three values based on the 30 years of data for that specific hour: a 10th percentile (only 10% of the total distribution is lower), an average, and a 90th percentile (90% of the total distribution is lower). Thus, for each three-hour game, there are nine potential values. To provide the broadest picture from the data, we plotted the lowest 10th percentile value, the highest 90th percentile value, and the middle average value. Thus, it’s quite possible to be hotter or colder on a given game day than shown here, just not as likely. Although humidity is another factor in human comfort and ball behavior, those data were not included in the hourly normals dataset and so are not presented here.
Cloud cover: NWS cloud cover data come as percentages of five cloudiness levels (clear, few, scattered, broken, and overcast) for any given hourly time slot; we used the values for the middle hour of any game for simplicity’s sake, as these don’t vary much on an hourly basis. Although precipitation can certainly affect fan comfort, those data were not provided in the hourly normals dataset and an initial look at the available raw data suggested that they would take more analysis than we could justify. Cloud cover can work as a very rough proxy for precipitation chances if desired.
Wind speed: Although a full exploration of this effect would require analyses of both wind speed and direction with reference to stadium orientation, the wind direction records in the hourly normals dataset are averaged vectors that do not present a clear picture of short-term variations, and acquiring and analyzing the complete raw dataset was more than we cared to do for this story. Thus we stuck with wind speed as a rough proxy for atmospheric conditions at the ballpark.
What month is your favorite at Kauffman Stadium? (No, wise guy, it’s not Always October)
This poll is closed