Hazy IPAs May Finally Kill Off the IBU

The measurement for hop bitterness in beer, the international bitterness unit (IBU), has always been problematic, but hazy IPAs may have broken it for good.

Let me take you back to a simpler time, fifty years ago. The triumph of lager was nearly complete. Sure, there were a few oddball beers being made in remote backwaters like Dublin (stout), Brussels (whatever it is Belgians make), and London (warm, flat bitter), but mostly the world had been made safe for proper beer—cold, delicious pale lager.* It has consequently been made safe for scientists, who enjoyed a study sample of a single beer style that encompassed most of the beer sold worldwide. In this environment was born a simple calculation that scientifically measured a subjective quality, bitterness, and had a handy scale that was as efficient and empirical as the thruster on an Apollo spaceship. The international bitterness unit (IBU) was born, and it promised a way to bring precision to a flavor quality a thousand years old.

IBUs are great when you can control for several major variables, as could lager-studying scientists of yore, but almost useless when you can’t. Critically, the perception of bitterness is profoundly affected by the type of beer in question. A strong, treacly barleywine can all but smother any perceptible trace of bitterness that 40 IBUs provide; yet that same measured bitterness in a helles would be overwhelming.

But there’s another, more pernicious problem. When conducting a test for IBUs, we test for an aggregate of more than one compound. As hops researcher Tom Shellhammer puts it: “International Bitterness Units are a chemical/instrumental measurement of the number of bittering compounds, specifically isomerized and oxidized alpha acids, polyphenols, and a few other select bittering chemicals, that make your beer taste bitter.”

And here we get to the increasingly problematic issue with IBUs in a post-hazy IPA world. Stan Hieronymus has another edition of his indispensable “Hop Query” newsletters out today (it’s free, and if you’re not a subscriber you are missing an amazing resource). He summarizes research by John Paul Maye at Hopsteiner, including this critical finding (my emphasis): 

The sensory bitterness of NEIPA is about half of what the IBU test result would indicate. This is due to the fact that alpha acids and humulinones interfere and absorb like isoalpha acids during the IBU test yet are only 1/10 and 2/3 as bitter as isoalpha acids. Decoded: a laboratory test measures a lot of “stuff” beyond iso. Hopsteiner did that test for each beer, but also calculated sensory bitterness as a sum of actual iso, two-thirds humulinones and one-tenth alpha acids. As a result, Beer B had a measured IBU or more than 70, and sensory bitterness less than 40. That was a pretty typical difference.”

The ascent of IBUs as an important stat for craft beers began a generation ago, as ultra-hoppy IPAs started making an appearance. Indeed, IBUs probably wouldn’t be much use to most drinkers except in the case of hoppy American ales (as a warning or lure, depending on the drinker). No one really much cares how many IBUs are in a saison or stout, so long as they’re a typical amount. (And if they’re atypical, how many drinkers know how to interpret the numbers, anyway? Are 30 IBUs too few in that stout; too many in that saison? What if that stout is a 4% Irish stout? Or a 6.5% sweet stout? Very few people have any idea what the number should be.)


It becomes more abstract when you throw hazies into the mix. Last night I was at Widmer and took note of their hazy offerings, surprised to see how high the listed IBU figures were. Widmer has a badass lab, so these weren’t expected IBUs—they were actual measured levels. Seeing how high they were—I think the first one was 70—I started ignoring them. While it’s great to have the data about why perceived and measured bitterness are out of alignment with hazies, I suspect many of us have already noticed this misalignment based on our own experience. This morning I posted a poll to see how others had been adapting to this change, and indeed, only about a third of you still pay much attention to IBUs. 

I wonder how much longer breweries are going to consider this a valuable piece of information to provide to customers?

Stan’s newsletter out a spotlight on another issue I’ve been pondering lately, though for the moment I’ll just bookmark this finding for later exploration. here’s Stan:

“The smooth bitterness that many people experience drinking NEIPAs is due to the high humulinone concentrations in these beers and low concentration of isoalpha acids. Again, you’ll find more about humulinones in the archive. In heavily dry hopped beers the concentration of isoalpha acids decreases as more hops are added and the concentration of humulinones increases.” [Note: humulinones are oxidative compounds in hops; more here.]

I definitely agree that the quality of bitterness differs if it’s derived from humulinones or iso-alpha acids. But I’m not one of the many who find that bitterness smooth. It’s less pleasant, more tannic, and less dynamic than standard hop (whole boil) bitterness, which I find more electric and lively, more a sensation than a flavor. To my palate, humulinones are more inert, dull, and grating. This is another way IBUs detract from our vocabulary rather than add to it—by collapsing iso-alpha acids and humulinones into one measurement, we lose the ability to distinguish them on our tongues. But let’s leave it there for now.


*This sentence written from the perspective of a brewing scientist in a white lab coat, circa 1968. 

Jeff Alworth5 Comments