Stadium Construction Trends in Baseball and Football

A short time ago, I authored a post at Beyond the Box Score about stadium construction trends in Major League Baseball. As promised, here's Part II of the series, focusing on pro football.

In this case, I started in 1940 in order to weed out (most of) the defunct NFL teams from the sport's early years, as well as some of the early sandlot fields for which it was near-impossible to uncover construction dates. Any remaining defunct teams are absent from the database. As with the last study, I counted shared stadiums only once.

There are some interesting parallels and contrasts between MLB and NFL/AFL stadium trends. Both sports exhibit two distinct construction booms, including an expansion-era boom during 1960-1975 and a modern boom 1995-present. But first, some fun facts:

• The average AFL/NFL stadium reached its peak age at the start of both the 1957 and 1958 seasons, at a mean of 36.4 years.
• As the NFL expansion era wrapped up in the mid-1970s, the average NFL stadium reached it's youngest point at 15.4 years of age in both 1975 and 1976.
• In the post-expansion years, stadium ages peaked again in 1991 and 1992 at a mean of 26.7 years. The latest wave of stadium construction drove that figure quickly down to post-expansion low of 15.5 years in 2003.
• That last wave of construction drove the average age down at a rate of nearly 1 year per season.
• The largest single-season "rejuvenation" took place from 1970-71. As Dallas, Philadelphia and New England all moved into brand-new stadia, and both Chicago and San Francisco moved into slightly newer facilities, the average age dropped by 7.5 years. Five seasons later, the average dropped by 5.0 years when Detroit, New Orleans and the NY Giants all relocated.
• The biggest single-season jump in age--a whopping 5.2 years--took place between 1941 and 1942, thanks to Philadelphia's flip-flopping between Shibe Park and a relatively new Philadelphia Municipal Stadium, and the (then) Cleveland Rams' similar vacillations between Cleveland Municipal Stadium and League Park.

As to compare NFL and MLB data, let's revisit the chart from Part I:

As noted above, the trends are very similar: two clear stadium construction booms occurring at essentially the same time. This shouldn't be too surprising, as the two sports commonly shared facilities up until the 1990s. What this suggests, however, is that stadium construction is driven as much by external (read: socioeconomic) factors as internal ones.

However, a closer look will reveal two key differences. First, the NFL trend is more volatile. As it were, NFL teams were far more likely to change stadia in their early years than were MLB teams. In later years, NFL teams also tended to play in multiple temporary facilities while awaiting construction for a new home. The NY Giants, who played in both the Yale Bowl and Shea Stadium while awaiting construction of Giants Stadium, are an example of this.

Second, the average MLB stadium has almost always been older than the average NFL stadium, sometimes by a full five years. See the chart comparing the two trends below.

The age difference is far more pronounced in the post-expansion era, but has persisted nonetheless.

This I was surprised to find, but I believe there are at least three factors involved here:

1. Professional football developed later than professional baseball.
2. Football teams were the early occupants of the multi-purpose stadium boom that characterized both sports' expansion eras.
3. Football expanded far faster than baseball.

Point #3 is key to the post-1960 data. While baseball expanded from 16 to 30 teams between 1961 and 1998, the NFL quadrupled in size from 1943 (when the Cleveland Rams took the season off due to WWII) to 2002. New teams usually mean new stadia, sooner or later.

That's it for this round of analysis. If you'd like to check out the raw data for the charts above, see the tables below.

Data Notes: There are three instances where a team split games between two official home fields. In 1995-96, both Seattle and St. Louis played half of a season in one facility while waiting for their new stadium to open. In 2005, New Orleans split their time between East Rutherford, Baton Rouge and San Antonio in the wake of Hurricane Katrina. In the first two instances I measured the age as the weighted average of the ages of the two stadia in which they played. In the Saints' case, I did the same but omitted their "home" game at Giants Stadium. Finally, sources disagree on where the 1945 Cleveland Rams played their home games, so I averaged the ages of Cleveland Municipal and League Park.

Year		MLB				NFL/AFL
		# Teams	Mean	YTY		# Teams	Mean	YTY
1940		16	28.4			9	24.2
1941		16	29.4	+1.0		9	24.4	+0.2
1942		16	30.4	+1.0		9	29.7	+5.2
1943		16	31.4	+1.0		8	30.4	+0.7
1944		16	32.4	+1.0		9	31.7	+1.3
1945		16	33.4	+1.0		9	31.5	-0.2
1946		16	34.4	+1.0		11	29.7	-1.8
1947		16	33.9	-0.5		11	30.7	+1.0
1948		16	34.9	+1.0		11	31.7	+1.0
1949		16	35.9	+1.0		11	32.7	+1.0
1950		16	36.9	+1.0		11	33.7	+1.0
1951		16	37.9	+1.0		11	34.7	+1.0
1952		16	38.9	+1.0		11	35.7	+1.0
1953		16	37.1	-1.7		12	33.9	-1.8
1954		16	34.7	-2.4		12	34.9	+1.0
1955		16	36.6	+1.8		12	35.9	+1.0
1956		16	37.6	+1.0		12	35.9	+0.0
1957		16	38.6	+1.0		12	34.3	-1.7
1958		16	37.0	-1.6		12	36.4	+2.2
1959		16	38.0	+1.0		12	36.4	-0.1
1960		16	37.3	-0.8		21	33.3	-3.1
1961		18	33.5	-3.8		22	29.5	-3.8
1962		20	31.9	-1.6		22	30.5	+1.0
1963		20	32.9	+1.0		22	33.7	+3.3
1964		20	31.0	-1.9		22	31.6	-2.1
1965		20	32.0	+1.0		22	32.2	+0.6
1966		20	30.7	-1.3		24	29.2	-3.1
1967		20	29.0	-1.7		26	27.7	-1.4
1968		20	27.3	-1.7		27	28.7	+1.0
1969		24	26.9	-0.4		27	28.1	-0.7
1970		24	24.7	-2.2		27	27.7	-0.4
1971		24	20.7	-4.0		27	20.2	-7.5
1972		24	21.5	+0.8		27	19.4	-0.8
1973		24	20.4	-1.1		27	19.3	-0.0
1974		24	20.1	-0.3		27	20.3	+1.0
1975		24	21.1	+1.0		27	15.4	-5.0
1976		24	21.2	+0.1		28	15.4	-0.0
1977		26	20.8	-0.4		28	16.4	+1.0
1978		26	21.8	+1.0		28	17.4	+1.0
1979		26	22.8	+1.0		28	18.4	+1.0
1980		26	23.8	+1.0		28	17.8	-0.5
1981		26	24.8	+1.0		28	18.8	+1.0
1982		26	24.8	+0.0		28	20.4	+1.6
1983		26	25.8	+1.0		28	21.4	+1.0
1984		26	26.8	+1.0		28	21.3	-0.1
1985		26	27.8	+1.0		28	22.3	+1.0
1986		26	28.8	+1.0		28	23.3	+1.0
1987		26	29.8	+1.0		28	22.4	-0.9
1988		26	30.8	+1.0		28	23.7	+1.3
1989		26	30.8	-0.1		28	24.7	+1.0
1990		26	31.8	+1.0		28	25.7	+1.0
1991		26	29.7	-2.1		28	26.7	+1.0
1992		26	29.0	-0.6		28	26.7	+0.0
1993		28	29.7	+0.7		28	27.7	+1.0
1994		28	27.4	-2.3		28	28.7	+1.0
1995		28	26.8	-0.7		30	28.5	-0.2
1996		28	27.8	+1.0		30	25.5	-3.0
1997		28	27.7	-0.1		30	25.2	-0.2
1998		30	27.0	-0.6		30	22.7	-2.5
1999		30	27.7	+0.6		31	22.3	-0.5
2000		30	22.8	-4.9		31	22.3	+0.0
2001		30	21.2	-1.6		31	20.5	-1.8
2002		30	22.2	+1.0		32	18.1	-2.4
2003		30	22.6	+0.5		32	15.5	-2.6
2004		30	20.8	-1.9		32	16.5	+1.0
2005		30	22.4	+1.6		32	18.2	+1.7
2006		30	22.1	-0.3		32	16.9	-1.2
2007		30	23.1	+1.0		32	17.9	+1.0
2008		30	22.4	-0.6		32	18.2	+0.2
2009		30	20.8	-1.6		32	17.9	-0.2
2010		30	20.9	+0.1		32	17.8	-0.1