But after anonymizing and compiling data from those probably-above-average-drinkers, the company released a report with a perhaps-surprising finding: People drink more in the winter. They created this graphic, an interactive version of which is at their site. The darker the color, the higher the average blood alcohol content for the day. (The drunkest day in February? Super Bowl Sunday.)
But while that might give you a picture of the time of year when Americans are drinking the most, not when they're getting into the most trouble. That yields a different results, looking at how often the phrase "alcohol appears to have been a factor" (or variants thereof) appear in the news.
I plugged versions of that phrase into Nexis, a database of news articles, to see how often it's used. Here's how often the term (again, terms, plural: "alcohol contributed," "alcohol was a factor," "alcohol may have contributed," etc.) have appeared each month since 1990.
The increase is due in part to the increased number of sources Nexis has in its database -- including the inevitable advent of Web-based publications. It may also in part be due to increased popularity of the expression itself. When you look at the average by month for all of those years -- which you see the opposite of BAC's drunkenness chart: There are more news articles about "alcohol being a factor" during the summer.
It still has bumps in the winter time and -- unsurprisingly -- in March (somewhere around the 17th, we assume).
But the peak is July. Since 1990, no month has been the month with the most "alcohol played a role" articles more often than that month. Because of the Fourth. Picnics. College kids home from school. And: People outside. Getting into fights.
Driving drunk. This isn't scientific, but when we search "alcohol was a factor," the next part of the sentence is usually "in the crash/incident/assault," etc. Warmer temperatures mean more crime. More crime means more crime where drinking was involved.