The Homogenization of Ballparks

Our goal is to design a world-class ballpark. That includes field dimensions that are entertaining for our fans and a playing field that’s fair to both pitchers and hitters,”
-David Montgomery, General Partner and President of the Phillies, April 2002 Press Release

Our No. 1 criteria was that we wanted it to play like the other Busch. We wanted it to play fair for the hitters and the pitchers.”
-Tony LaRussa, manager of the St. Louis Cardinals in 2009, on Busch Stadium III

First and foremost, we wanted to have dimensions that played fairly.”
-Erik Judson, San Diego Padres Vice President of Development when Petco Park was built

That’s a lot of lip service to creating ballparks that play fair, all specifically in reference to ballparks built in the last 20 years. You surely could find similar quotes about most ballparks built in that time frame. Implicit in those statements is that ballparks that came before possessed different dimensions that didn’t play fair, at least not fair to any sort of league-wide standard. There’s a lot of truth to that notion. Ballparks today have become more and more homogenized, while ballparks of yore possessed wild fluctuations in dimensions.

No ballpark better encapsulated these fluctuations than the Polo Grounds. For the last 41 years in which baseball was played at the Polo Grounds, the distance to straight-away center ranged from 475 to 505 feet away from home plate. By contrast, the distance to the foul pole in left field was a mere 280 feet, and right field was a miniscule 256 feet. It was a horseshoe.

Those distances are eye-popping because nothing like that exists today. The Polo Grounds were not alone in their oddity. These types of idiosyncrasies have disappeared from the game.

Let’s take a deeper look. We’ll look at distances to the outfield fence, the area behind home plate, ballpark capacity, the height of the fence, and discuss the way the two biggest ballpark booms in the history of the game have shifted the story.

Outfield Distances

This all began as an effort to see exactly how ballparks have changed over time. Using the incredible ballpark database at Seamheads (data through 2015), I was able to collect year-by-year data for every ballpark. While some of the data were incomplete going back into the 19th century, there was ample info for outfield distances to the left field pole, dead center field, and the right field pole going back to 1895. One of the easiest ways to approach this is to look at the maximum, minimum, and average distances for the two foul poles and dead center field.

The averages themselves have remained mostly flat, especially in the corners. Since 1947, the gap between the longest average distance to left field and the shortest distance is a tine six feet. You have to go all the way back to 1899 to get a gap between largest and smallest of more than 20 feet in right field . Center field has had more variance, but even its top and bottom averages are within 10 feet of one another since 1965.

These graphs are a good start in illustrating the variance between the maximum and minimum distances, but it only accentuates the outliers. Let’s approach it from a different angle, with a bubble graph illustrating every single ballpark dimension since 1895. To illustrate the homogenization over time, I’ve added some transparency to the bubbles. The hotter the color, the more bunched the dimensions are, meaning there’s more similarity across all ballparks.

Now we’re really starting to see the homogenization amongst outfield distances. You can see the red bubbles beginning to bunch in the early 1960s, and the color only starts to become more intense from there. And it’s true of all three outfield locations. The one outlier since 1981 is the center field distance at Minute Maid Park, which has been 436 feet from home plate since opening for play in 2000. However, even Minute Maid will undergo a renovation that diminishes the distance to a more standard 409 feet in time for the 2017 season.

Last, but not least, let’s take a look at the size of a single standard deviation by year.

Variance has been on a steady decline since the very first game in 1895 other than a few spikes. Usually those spikes were caused by teams moving into temporary homes. For instance, the mid-1990s mild spike in right field was caused by the Rockies’ brief two-year turn in Mile High Stadium while waiting for the completion of Coors Field. A similar skewing happened when the Dodgers called the Los Angeles Memorial Coliseum home in the early 1960s.

And it’s also important to mention that there are more and more ballparks in the sample each year. The 1895 sample includes 11 ballparks, 1919 has 15, and more modern numbers feature 30 or more. A larger outlier in the earlier years will have a more dramatic effect.

It’s plain to see that ballparks are becoming more and more similar across outfield dimensions. I’m not showing my math here, but I did look at data for the power alleys–left-center field and right-center field–and the results were almost completely identical. Homogenization was already starting to happen in the 1940s and 1950s. Then the Concrete Donut boom (1965 to 1971) sent homogenization into hyperdrive. Twelve new ballparks entered the game in that time frame (highlighted on all of these graphs), and many of them were multi-purpose, cookie-cutter affairs. The collective effect was the collapse in the number of idiosyncrasies among dimensions.

Behind Home Plate

Here, I’ve combined the maximum, minimum, and average graph with the standard deviation, placed side by side.

The maximum/minimum graph is not particularly illustrative because the two most prominent maximum distances, belonging to old Comiskey Park and Yankee Stadium, remained unchanged for the majority of their existences. Even as other ballparks with larger areas behind home plate were replaced with new ballparks with smaller distances, those two ballparks held on. It wasn’t until Yankee Stadium was replaced after 2008 that we really see the gap between the maximum and minimum collapse.

Over the last eight seasons, that gap is lower than ever before. The same holds true for the standard deviation, which has totally disappeared. It stands at a miniscule 4.88 feet at this point. The deviation was never truly that large to begin with, but now it borders on non-existence. Like the distances to outfield fences, the idiosyncrasy in another ballpark aspect has disappeared.

Capacity

As I did with the area behind home plate, I’ve combined the seating capacity maximum, minimum, and average graph with the standard deviation, placed side by side.

These graphs are a bit noisier. Increased interest in the game muddies the story. First, radio became more prevalent and bumped the game’s standing in the national psyche. Then television pushed it to further heights. And while all of this was going on, teams were constantly moving into new ballparks, tweaking the numbers in one way or another. This is most obvious in the early 1960s, where the cavernous 92,000-seat Los Angeles Memorial Coliseum yet again rears its head. It’s also apparent that 1933 to 1956 was an era in which very few new ballparks began use. It stands out as one of the few truly steady eras in the standard deviation.

Still, there’s a clear pattern of homogenization in capacity after the Los Angeles Memorial Coliseum and Los Angeles’ Wrigley Field (the initial home of the Angels) were vacated. Yet again, the Concrete Donut boom had a profound impact on the size of deviation. Then the second major boom–we’ll call it the Camden Boom–finished the job. From 1989 to 2012, there were just six seasons in which a new ballpark didn’t open.

This homogenization in capacity is at least in part due to better marketing. Emptier stadiums look worse on television, and fans in attendance have a better time watching a good or even average team when the house is full. Teams realized that it’s better to sell out a 38,000-seat stadium than it is to place those same 38,000 fans in a 55,000-seat concrete donut. Sure enough, the average capacity has decreased from a peak over 50,000 in 1993 to a much more realistic 43,218, with most new ballparks built in the Camden Boom pushing the average incrementally lower.

Admittedly, the deviation gets noisy in the early 2000s, but that is almost exclusively a result of the Expos’ presence at the 18,000-seat Hiram Bithorn Stadium prior to their move to Washington DC. Note that in the interest of the spirit of the graph, Hiram Bithorn was removed from the max/min graph, as it was not a stadium built with the intention of housing an MLB team. Rather, it was a less-than-full-season temporary solution.

Wall Height

Finally, let’s review the height of the wall. This data is slightly less robust. It only reliably appears since 1900, so we won’t see the 1895 to 1899 seasons here. Since wall heights have constantly been affected by outliers throughout history (the Green Monster says hello), I’ve opted to skip the maximum/minimum graph. Instead, we’ll use the transparent bubble treatment again.

Here we can see that Fenway is an outlier in left field and remained as such for the majority of the sample. But more importantly, we finally see an aspect in which homogenization has reversed itself. Franchises have decided that wildly variable wall heights are an acceptable way to make a ballpark idiosyncratic.

There was a brief period in which wall heights became homogenized after the Concrete Donut boom, but it looks like the Camden Boom corrected it to some degree. Jacobs/Progressive Field and Minute Maid Park stand out as ballparks that entered play during the Camden Boom and employed mini-Monsters in left field. Chase Field has a 25-foot wall in center field, and Marlins Park opened in recent years with a 20-foot wall in center field.

The Metrodome’s right field garbage bag held up the standard for unique right field walls in the 1980s at 23 feet high. Then the Twins carried over the outlier into the new Target Field. By that time, AT&T Park, PNC Park, and Camden Yards–three of the biggest jewels of the Camden Boom–had joined in the fun of an idiosyncratic right field wall.

To drive home the point, here are the standard deviations since 1900 for wall heights.

First, it’s worth addressing the giant spike in the early 2000s in center field. That was a result of the Reds’ tweak to Cinergy Field, which goosed the center field wall to 40 feet high. It was done to protect the integrity of the batter’s eye while construction of Great American Ballpark took place in plain sight. Left field spiked in the late 1950s–there’s the Los Angeles Coliseum again, this time with a 40-foot wall in left field–but the standard deviation for left field wall heights died during the Concrete Donut Boom and never recovered.

Where it gets fun is in center field and right field during the Camden Boom. Seemingly, teams with new ballparks wanted to be unique, but without overtly copying Fenway’s famed Green Monster and without employing the cartoonish distance dimensions of the past. The answer was a bump in wall heights in center and right fields, which is reflected in the increase in standard deviation during that time frame.

The other way franchises have fought total homogenization is with off-field thumbprints. Camden Yards has its warehouse. Marlins Park has an aquarium behind home plate and bizarre moving “artwork” beyond the left-center field wall. Minute Maid Park has a miniature train that runs along the exterior left field wall, and AT&T Park has McCovey’s Cove, amongst so many others. It’s not that ballparks have become entirely homogenized. Only the on-field aspects have become more and more vanilla.

References & Resources

• Seamheads, Ballparks Database
• Philadelphia Phillies, Press Release, “Phillies reveal New Ballpark field dimensions”
• George Vecsey, The New York Times, “Glorious Stadium Honors a Glorious Past”
• Corey Brock, MLB.com, “With subtle changes, Padres hope Petco plays fair”
• Ryan Dunsmore, The Crawfish Boxes, “Astros confirm center field renovation before the 2017 season”
• Chicago Tribune, “Reds’ Center-field Wall Will Top Green Monster”