What do you think of when you think of a “summer” beer?
For me, it’s usually either Pyramid’s Curveball Golden Ale or Deschutes’ Twilight Ale. The former holds a special place in my heart – it’s what was served at my wedding – while the latter, mostly due to a combination of price and ubiquity in grocery stores ‘round these parts, has become my go-to six-pack to pick up on my way to summer gatherings. The two share two notable qualities, though, which I’d like to focus on here: they’re available only during summer, and they’re listed by their makers as golden ales.
Although I technically don’t have the data to back this up, I think it’s safe to say those two brews are in line with most summer seasonals when it comes to style, or at least general “color.” Not a lot of breweries are putting out spiced heavy ales once June rolls around – it’s all goldens, blondes, wheats, and pales. And this mostly makes sense, when you think about it – warmer temperatures and sunny days make people want something lighter and more refreshing. Right?
To examine this question, I looked at ratings by beer style aggregated in two separated time periods, January/February (i.e., Winter) and June/July/August (i.e., Summer), and checked how they compared. Because I did this by style and not by individual beer, the ideas of replacement level, wOBAR, and BAR didn’t really apply – all the numbers you see in this article are raw ratings only. To be included in the sample, a style had to have at least one check-in in every month in the set; this excluded three, leaving 147 for comparison, with check-in totals across all five months ranging from 674,000 (okay, actually 674,001) for the bread-and-butter of Untappd’s craft-drinking user base, the American IPA, to a mere 23 for Weissbier.
The first major result I found is that is that Untappd users like beer more in the summer, though not in August. The average rating in winter months comes in at 3.09, while in summer it’s up to 3.17 (or 2.8%).
Now, the astute reader may notice two things about the graph above. First, the vertical scale doesn’t start at zero, but instead at 2.9; this is to better illustrate the month-to-month and season-to-season difference. Second, and more significantly, I didn’t include any error bars on the graphs. I did determine the standard deviation for all the values I discuss in this post, and for any and all beers with a decently-sized check-in volume (at least 100 or so) the SD falls between 1.2 and 1.8. Fortunately, I was able to confirm statistical significance of the change (and the season-to-season change of every style I’ll mention below) using a t-test, so I feel okay about leaving them off for the sake of prettiness of the graphs.
So, armed with the knowledge that, in general, beers are rated about 3% higher in the summer, let’s see how individual styles fare. We’ll do that by examining how much higher each style rates in the summer relative to how much higher the entirety of the sample rates excluding the style in question (to eliminate any effects the more popular styles could have on the average). In other words, we’ll look at
and see if breweries have the idea of a summer-release beer right, or if they should try something else.
Now, as mentioned before, there are 147 styles being included here. This leads to some potentially more obscure styles near the top and bottom of the list, which can potentially obfuscate the results. For example, the absolute highest jump from winter to summer belongs to the Burton Ale style, at +21.8%, which I’d never heard of before starting the research for this post. However, excluding any styles with fewer than 5000 check-ins (or about 0.1% of the total sample) eliminates most of these, and although I wouldn’t want to do that when getting the quantitative results, it can make finding points of discussion easier.
After the trimming, the “Radler” style jumps to the top of the list, seeing about a 12% increase in rating in the summer. If the name is unfamiliar, perhaps you know it instead as a “shandy” – this style is a roughly 50:50 mix of beer and soft drink (in the U.S., most commonly lemonade). Wisconsin’s Leinenkugel Brewing Company recently went national with their distribution – perhaps you’ve seen their Summer Shandy in your local store? That’s the sort of thing we’re talking about here. These typically are summer-release only, and it certainly seems the brewers are making the right call there. Malt liquor sees the second-highest jump, at about 9%, which seems related to the half-lemonade beer – the malt liquor sample is mostly made up of assorted Mike’s Hard beverages.
Others near the top aren’t too surprising – Blonde Lagers, Pale Lagers, and Helles Lagers all see a significant increase over expectations, as well as the various pilseners, all of which match with the idea of summer beers being lighter. Even American adjunct lagers (i.e., Bud, Coors, etc) get in on the phenomenon. In fact, lagers in general get a greater boost than ales, though the difference is small (3.1% versus 1.9%).
Three other notable styles stand out at the top: smoked beers/rauchbiers, English strong ales, and winter ales. I can sort of rationalize the smoked beers – maybe people thing they go well with grilling/barbecuing? – but I don’t know what to say about the strong ales and winter ales. One possibility comes from the fact that I didn’t filter by check-in country at all, so perhaps the winter ale check-ins are coming from the southern hemisphere, and thus would be reversed relative to my expectations?
Regardless of the reasons, one thing definitely jumps out about the other end of the list: six of the eight worst summer beers are “imperial” beers of one type or another. Two even see a negative change from winter to summer, not just a below-average gain – imperial oatmeal stouts and imperial red ales. Even Imperial IPAs, which have the highest absolute scores in both time categories, see a summertime score rise well below the average. It would seem that in the summer months, people prefer a beer with less alcohol.
So, what conclusions can be drawn from all this? Primarily, it seems to me that breweries are actually getting it right. With a few exceptions on both the high and low ends, the beer styles that see the highest boost in the summer, relative to average, fit the mold of the traditional summer seasonal, at least in terms of malt color. The data also suggest that it might be wise for brewers to increase their lager offerings in the summer, as they generally see a little bit more of a boost than ales. Perhaps that’s something we’ll see in future summers.
The full data list (with the rarer styles pushed to the bottom):
|Style Name||Winter Score||Summer Score||% Boost over Average||Check-in %|
|American Amber / Red Lager||2.794||3.013||5.00%||0.84%|
|English Strong Ale||3.014||3.246||4.80%||0.20%|
|Japanese Rice Lager||2.503||2.687||4.45%||0.19%|
|Canadian Adjunct Lager||2.376||2.546||4.30%||0.21%|
|Extra Special/Strong Bitter||2.911||3.113||4.09%||0.63%|
|Foreign / Export Stout||3.114||3.328||4.01%||0.14%|
|American Brown Ale||2.977||3.169||3.66%||1.53%|
|American Adjunct Lager||2.381||2.538||3.66%||2.92%|
|Kellerbier / Zwickelbier||2.863||3.048||3.64%||0.33%|
|American Blonde Ale||2.723||2.897||3.54%||0.56%|
|American Amber / Red Ale||2.909||3.091||3.48%||2.54%|
|English Pale Ale||2.880||3.057||3.33%||1.15%|
|Scotch Ale / Wee Heavy||3.189||3.378||3.11%||0.75%|
|Milk / Sweet Stout||3.359||3.545||2.66%||0.97%|
|Dunkel Munich Lager||2.918||3.076||2.60%||0.19%|
|American Pale Ale||3.054||3.215||2.60%||5.77%|
|English Brown Ale||2.953||3.111||2.56%||0.62%|
|Belgian Strong Dark Ale||3.243||3.412||2.38%||1.27%|
|American Strong Ale||3.314||3.483||2.29%||0.98%|
|Black IPA / Cascadian Dark Ale||3.183||3.340||2.15%||1.10%|
|Russian Imperial Stout||3.479||3.652||2.11%||1.20%|
|Belgian Pale Ale||2.975||3.112||1.83%||1.01%|
|Irish Red Ale||2.891||3.021||1.71%||0.57%|
|Dortmunder / Export Lager||2.777||2.902||1.69%||0.17%|
|Belgian Strong Pale Ale||3.181||3.321||1.62%||0.83%|
|Scottish Export Ale||3.174||3.314||1.61%||0.11%|
|Flanders Oud Bruin||3.232||3.375||1.61%||0.17%|
|Gruit / Ancient Herbed Ale||2.935||3.061||1.53%||0.11%|
|Saison / Farmhouse Ale||3.062||3.192||1.47%||2.36%|
|American Light Lager||2.248||2.346||1.45%||2.57%|
|Imperial / Double Black IPA||3.447||3.587||1.24%||0.24%|
|Irish Dry Stout||3.348||3.482||1.19%||0.77%|
|American Pale Wheat Ale||3.001||3.119||1.13%||1.98%|
|English Mild Ale||2.881||2.993||1.09%||0.16%|
|Imperial / Double Porter||3.411||3.530||0.68%||0.43%|
|Biere de Garde||2.999||3.098||0.54%||0.19%|
|Flanders Red Ale||3.441||3.544||0.22%||0.30%|
|Imperial / Double Pilsner||3.035||3.125||0.18%||0.20%|
|Belgian Brown Ale||3.036||3.123||0.09%||0.13%|
|Belgian Blonde / Golden||3.133||3.209||-0.32%||0.60%|
|Imperial Pale Ale||3.204||3.281||-0.33%||0.40%|
|American Imperial / Double Stout||3.546||3.641||-0.37%||2.57%|
|American Wild Ale||3.335||3.406||-0.64%||0.65%|
|Imperial / Double Brown Ale||3.352||3.385||-1.75%||0.14%|
|Imperial / Double IPA||3.653||3.664||-2.88%||7.08%|
|Imperial Oatmeal Stout||3.483||3.410||-4.74%||0.14%|
|Imperial / Double Red Ale||3.640||3.450||-7.87%||0.71%|
|Euro Dark Lager||2.693||2.996||8.29%||0.09%|
|Black & Tan||2.837||3.155||8.26%||0.08%|
|English Ale (Decommissioned)||2.697||2.966||7.02%||0.02%|
|Hefeweizen Light / Leicht||2.537||2.745||5.30%||0.01%|
|Biere de Champagne / Biere Brut||3.182||3.294||0.74%||0.02%|
|Belgian Trappist Ale||3.125||3.205||-0.18%||0.00%|
|New Zealand Pale Ale||3.161||3.238||-0.31%||0.08%|
|Australian Pale Ale||3.169||3.197||-1.81%||0.08%|
|Belgian Ale (Decommissioned)||3.276||3.281||-2.52%||0.00%|
|Australian Sparkling Ale||2.897||2.826||-5.08%||0.01%|