An Impromptu Low-ABV Analysis

Nathan Gismot, June 26, 2014

Lately, I’ve been having peculiar BeerGraphs daydreams with a nautical motif. In them, I imagine myself taking an heroic swan dive into our Leaderboards in search of knowledge-treasure. I imagine myself swimming ever deeper into the briny depths, and that the dulcet tones of a siren song entice me onward. A bright ring of coral appears on the ocean floor below me, and I discover within it a strange and wondrous (if sudsy) pearl of evidence-based wisdom. I cradle the newfound knowledge-nugget in my outstretched hands and, emerging triumphantly from the deep, distribute its spoils to you, the enlightened People of BeerGraphs.

Delusional ravings aside, I am here to report I have quite failed in my research thus far to even scrape the surface of that which is epic and pearly and opalescent. Instead, I have scraped the surface of that which is weak and curious and questionable.

That is to say, I have conducted a brief research of low-ABV beers.

“Low-ABV beers,” for the purposes of this article, are defined as falling somewhere between 1.0% and 3.0% alcohol-by-volume (ABV). As I conducted my search, I left the BAR filter wide open (-15.0 minimum to 19.0 maximum), left Popularity untouched at its “All beers” default, and decided against filtering by region. (Hardly worthy of a dip in the pond, let alone a deep-sea dive, but here we are.)

A few words before I discuss my initial peek at the results.

BeerGraphs has featured a number of excellent articles concerning ABV, including (but certainly not limited to) Matt Murphy’s “Some Like it Hot: How ABV Influences Ratings,” Miles Liebtag’s “High Functioning: Craft Beer and Alcohol,” and Alex Fossi’s “Are Stronger Stouts Better Stouts?

One of my takeaways from those articles is that there exists a roughly positive correlation between ABV and ratings – to a point. Indeed, as Matt demonstrated in “Some Like it Hot,” different styles have different ratings “peaks” where ABV is concerned; and, once ABV continues to increase beyond that “peak,” the ratings tend to decrease slightly, indicating (hey, presto!) a negative correlation.

That said, those data nevertheless support the notion that high-ABV beers are generally rated better than their low-ABV counterparts within their respective style classifications. Miles, in the article I linked to above, makes an observation that further supports that conclusion: “A quick look at RateBeer’s Top 50 bears this out: of the 50 beers there listed, only 4 of them are under 8% alcohol by volume…the other 46 range from 8% on the low end to 15% and above, depending on vintage and release.” 

Data notwithstanding, common sense would seem to agree that consumers – especially ones who are serious enough about their beer to seek and create information about them on platforms such as Untapp’d – tend to prefer beers with higher ABVs. Alcohol is part of the experience, after all.

Let’s turn back to my little search. Duly girded with my piddling understanding of my colleagues’ research and, of course, my assumptions, I futzed with the Leaderboard filters as described above and hit “Enter,” fully anticipating a carnival of terrors: all BARs in the tank, Style+ way under the century mark, etc.


139 beers were returned in my search, and I sorted them in descending order by Style+. A full 31 beers were at  or above a 100 Style+, and another 31 rate between 99 and 95. What’s more, I had to scroll down to the 86th beer on the list – Blitz Cherry by BrewDog – before Style+ began to slip under 90. (For the morbidly curious among you, Skol Lager by Carlsberg checked in at dead last [by Style+] with a 61.)

Quite literally, that’s not half bad. What’s going on here?

It seems pretty straightforward, actually. For one thing, many, if not most beers on the list appear to be brewed and consumed entirely outside the United States. In other words, the data may be unduly influenced by limited distribution and small sample sizes. If this were a proper research project, which it most certainly is not, then I would gather raw check-in data for each of the beers, display it for you in a table, and then analyze those data, controlling for sample size and other fun variables.

Alas, dear reader. You will have to take me at my word that I poked through Untapp’d for the first handful of beers on the list and noticed that most of them have low check-in numbers. Take that for what it’s worth.

Another consideration is that many/most of the beers on the list are classified as Berlinner Weisses, which tend to have low ABVs on the whole. It seems at least feasible – if not likely – that critical consumers who were interested in drinking low-ABV beers might be drawn to the Berlinner Weiss style and, therefore, rate them more favorably than they otherwise would.

This leads me to my final thoughts: just because ratings generally improve as ABV increases doesn’t mean there aren’t consumers who prefer low-ABV beers, whether situationally or across the board. I, for one, can recall a number of occasions on which I selected a low-ABV beer simply because I didn’t want to get walloped with alcohol for one reason or another. In those instances, my decision point stood on the convergence of two factors: 1. What was available; 2. Which of those available beers was pretty good, as per my palate. 

I’m interested in outliers, and so I think there’s another research question in here somewhere. I’m not above pandering: what do you think about low-ABV beers? Are you an enthusiast? Do you know an enthusiast? I’d love to learn more.

Hmm. I might get to take that little nautical-themed daytrip after all. 

You can follow Nate on Twitter & Untapp'd @nategismot.