This is an infographic studying traffic fatalities during NFL gamedays, comparing when the home team is in town vs when they aren't.

Researchers looked at the last 10 years of these games, studying traffic fatalities that occurred within one hour of the stadium.

What they found was interesting--in most cases, pro football doesn't seem to move the fatality needle in a significant way--but in some cities (Dallas and Seattle), the roads get less safe during football games.

While in other cities, like Denver, New England, and Tampa Bay, the streets are actually safer during them!


Infographic: NFL Games Make Some City Streets Safer, Others More Fatal

In a staggering new infographic, we learn that NFL home games make for pretty safe streets—except in Dallas and Seattle.

When most people think NFL safety, they think concussions. But this pro sport can actually affect your well-being off the field. Because while it's safer to drive in Denver when the Broncos are in town, if you’re a local Cowboys or Seahawks fan, you might be safer steering clear of the stadium.

Why? John Nelson, co-creator of the "Game-Day Traffic Fatalities" infographic you see here, has no clue. His graphic breaks down traffic fatalities over the last decade of NFL games. He looked only at areas within one hour of the stadium, and he only counted game days. And after digging through all of this information, he discovered something incredibly interesting: In most cities, a football game doesn’t seem to affect traffic safety. But in some cities, pro football could make the roads significantly more dangerous (Dallas and Seattle), and in some cities, it could make the roads significantly less dangerous (Denver, New England, and Tampa Bay).

"Overall, I'm nervous," Nelson admits to me. "This is the sort of analysis you especially don't want to get wrong."

That said, assuming his methodology was as flawless as it appears to be, the findings are incredible. Looking at the image, you can see the percentage of traffic fatalities on game days, when local team was home versus away. So far, so good? Okay. Realize that most of what you’re looking at isn’t important—it’s not statistically significant, which probably means NFL activity didn't generally affect street safety in most cities. Now look for which numbers have boxes around them. Those are statistically significant; they’re the outliers that might teach us something about what some cities are doing right and others are doing wrong.

To put that potential wrong into perspective, in Dallas, there were 6,052 traffic fatalities from 2001 through 2011. 65 traffic fatalities happened during game days. 47 happened on days when the team was playing in town, and 18 happened when the team was playing out of town. Put differently, about 2.5x more traffic fatalities occurred in Dallas when the Cowboys were in town.

Yet Nelson isn't focusing on the negative, because he'd expected drinking fans and more traffic to drive a stronger trend of fatalities. "Honestly, I was surprised more by the venues that appear to be safer when the team is playing in town," Nelson tells me. "But after more thought I can offer hunches as to why that is: Slower, more carefully choreographed, traffic. Tens of thousands of cars not driving when the game is on. And tailgating cultures that extend that period of non-driving."

It’s entirely possible, Nelson reasons, that tailgating culture could actually make roads safer. But ultimately, we just don’t know. Now that this information is out there and laid out with relative clarity, hopefully cities like Dallas will do some introspection, while cities like Denver can become models for everyone else.

See more here.

Add New Comment


  • John McHugh

    Not only is this incomplete as others have pointed out, but it's a prime example of the misunderstanding between causation and correlation. But the kicker for me is the lack of a succinct point of view or conclusion. Pretty much negates the point of data visualization.

  • Mark Wilson

    I think we're getting into the idea of, should data be visualized if we don't fully understand it, and if we can't draw one strong conclusion from it. And should data be visualized when it doesn't necessarily matter--when it's not significant. Is it context or superfluousness? 

    I also don't think the viz authors are making any claims of a direct causation--more that this often meaningless but sometimes quite strong correlation deserves exploration as to its causation. 

  • myrna652

    as Kathy responded I'm taken by surprise that a student can make $4758 in 4 weeks on the computer. here are the findings w­w­w.J­A­M­20.c­o­m

  • MyNameIsMattRogers

    Looks like San Francisco and Oakland are grossly mis-represented here though...

  • MyNameIsMattRogers

    "Where are the Bay Area, NYC, and DC teams? The nature of this map looks at an individual team's local area.  When local areas overlap with other local areas it muddles the home and away differentiation.  So they could not be considered in this map."
    Looks like they bundled Baltimore and Washington in with Philly.