As a result of Linsanity, people that had little to no interest in basketball became fans of the Knicks and, more importantly, fans of NBA basketball in general. I couldn’t give two craps about the “real” fan aspect during a time of when the Knicks were in dire need of a point guard that wasn’t an injured, decrepit Baron Davis. And then, when Lin left, (I’m assuming) fans of Lin and the Knicks shifted over to Houston, because their basketball hero went elsewhere, prompting people calling Knick fans some of the worst fans in the league. But, however, Emory University’s sports marketing analytics department disagree based off of the field of finances, at least.
Where a game takes place hugely impacts performance, without even taking home field advantage into account. In the MLB, there are “ballpark factors” which provide data as to how much more or less likely an event (e.g. double, home run, etc.) is in a particular ballpark relative to the league average. These “ballpark factors” are a concept that most die-hard baseball fans know very well, and something all fantasy baseball players should be familiar with (especially for daily sites, like FanDuel). ESPN provides a table to all its readers like the one shown on the left. The table reads as follows: for every one run scored in a league average MLB park, 1.501 runs will be scored in Colorado and .825 runs will be scored in Seattle. These factors are not the be-all and end-all when it comes to explaining player performance, but it’s another predictive tool to add to your tool belt.
Two recent examples show its application quite nicely: the performance decline of Robinson Cano after he decided to move to Seattle this past season, and the reasoning (or lack thereof) behind the Mets free agent signing of Michael Cuddyer this offseason. In his seven full seasons with the Yankees, Cano averaged 24 home runs in 160 games per year. His first year in Seattle he hit 14 in 157 games. Using only ballpark factors we would have predicted that he would hit 16 – not bad for just one calculation. Michael Cuddyer has played his last three seasons with Colorado, hitting .307 with 15 home runs in only 93 games per year. Again using only ballpark factors, in 93 games next year he should hit .254 with 12 home runs.
Being the fantasy sports aficionado that I am, I wanted to apply the same idea behind these ballpark factors to NFL data. However, much to my dismay, there was no NFL equivalent to be found. So, I decided to create NFL Stadium factors based on data from 2010 to 2013. The result is the table seen below.
Just as with MLB “ballpark factors”, the numbers in this table are just another piece in the puzzle of football analytics. Unlike baseball, however, the NFL Stadium Factors are a slightly more effective tool on a team-by-team basis rather than for individual performance. For example, consider the trade rumors hovering around the weeks leading up the NFL draft this past year. “Brady to Houston,” the headlines read. On the surface this looks like a no-doubter for the Texans, but has New England’s stadium been augmenting Tom Brady’s statistics over these many years? In fact, a quick look at the table on the left makes me wonder if the Patriot offensive juggernaut as a whole has benefited by playing in Foxboro. New England’s passing attack has averaged 276 yards per game and 33 touchdowns in the air over the past five years. If they had been playing in Houston’s stadium over that time span, stadium factors suggest those number would have plummeted to 252 and 29.
It’s easy to take these numbers as they are and just plug them in your statistical analyses; however considering the characteristics of given stadiums in order to understand why certain trends exist in the data is infinitely more useful. The stadium characteristics I looked at are as follows: domes, turf fields, cold weather, noise, and altitude.
- In stadiums with domes, you see a significant increase in field goals made and a decrease in rush yards gained. This is likely due to the absence of wind and other adverse weather effects, which negatively affects the passing game and field goals. Regardless of whether or not a team has a good rushing attack, if conditions don’t lend themselves to a game plan centered on throwing the ball, then more rush yards will inevitably be gained.
- On turf fields, the number of successful field goals goes up substantially due to the more consistent footing for kickers. Kickers are much more prone to slipping on the less-secure grass footing of a natural surface.
- The third characteristic, cold weather, is defined from a list of the 10 coldest and snowiest stadiums in the NFL. From that classification, I found that noticeably fewer points, rush touchdowns, and field goals occur in those stadiums. These outcomes are all fairly logical and can be explained by the unfavorable effects of the cold on the human body. In addition, as the temperature drops the football becomes less elastic. In combination with the dense, cold air inside and outside of the football, this makes field goals a much harder task.
- In the NFL’s five loudest stadiums, noise was found to lead to fewer points scored per game, most likely because of a two factors – communication and intensity. The louder a stadium gets, the harder it is for an offense to communicate certain blitz protections or other audibles. Secondly, the intensity of a loud crowd leads to more pressure and greater nervousness, which I believe more heavily impacts offensive performance.
- Finally, altitude is directly and positively correlated to field goals made and rush touchdowns scored. The first part makes perfect sense – things fly further in the lighter air of high altitude. The reason behind the second finding is a little more intricate. Altitude has powerful effects on lung capacity and conditioning levels, so defensive linemen (who aren’t in shape) tend to struggle in places like Mile High Stadium. Rushing touchdown data would specifically reveal this trend, because they often occur at the end of long drives when those in charge of stopping the run (the defensive linemen) are exhausted.
This table is a great starting point in starting to describe the effect that a certain location has on NFL performance. Although the insights behind the aforementioned explanations are my personal opinions, the numbers can be explained logically, and when used in statistical analysis will most definitely lead to improved results.
Michael Byman (@MichaelByman) is a senior at Emory’s Goizueta Business School. He is a Sport Analytics Research Grant recipient & submarine college pitcher.