Jason A. Hoffman

Jason A. Hoffman, PhD

You may have this same division among your friends and family. At my table it runs between Diet Coke and Coke Zero: a friend drinks the first, I drink the second, and both of us believe the science is on our side. So I went and read the science. A disclosure before the findings: I hold a long position in Coke Zero, chilled, ideally from a can, with one exception I will come to. What follows is nevertheless a fair summary of the evidence.

The first finding is that these are not two versions of one drink. They are two different drinks. Diet Coke, launched in 1982, was never designed to taste like Coca-Cola. It is its own formula with its own flavor system, sweetened with aspartame, brighter and sharper than the flagship. Coke Zero, launched in 2005, received the opposite brief: approximate the taste of Coca-Cola as closely as possible without sugar, which is why it runs a blend of aspartame and acesulfame potassium and a flavor base aimed at the original. Preferring Diet Coke is therefore not a statement about cola at all. It is a statement about Diet Coke.

The second finding is that the taste-test literature cannot crown a winner, because taste tests do not measure what people think they measure. The Pepsi Challenge showed that a blind sip favors Pepsi, while the market of its era kept favoring Coke: a sip is not a habit. New Coke is the canonical case. In 1985, Coca-Cola reformulated after blind tests said the new formula beat both the old one and Pepsi. The tests were real, and the market revolted anyway; the original returned seventy-nine days later. Preference turned out to be attached to the brand, the memory, and the ritual, not the liquid. In wine studies, the same wine is rated better when poured from an expensive-looking bottle. Expectation is an ingredient.

The third finding is mere exposure: repeated contact with a thing makes you like it more. Your cola preference was mostly installed by whatever you were drinking when you were becoming yourself. It is not a conclusion. It is an archive.

So what does preferring Diet Coke say about you? That you are loyal to a beverage that never pretended to be anything else, which is a coherent and even admirable position. Also that you drink a brand rather than a flavor: you have opinions about cans versus bottles, you believe the fountain version at McDonald’s is objectively superior, and you regard reformulation as a category of betrayal. You did not choose Diet Coke after comparing it to Coke Zero. You chose it in the era when Diet Coke was an identity, and you kept the faith. There are worse things to be than faithful.

One more disclosure, because the Diet Coke person’s strongest belief turns out to be simply correct. The McDonald’s exception is real, and I hold it myself. McDonald’s chills the syrup and the lines, filters the water, sets the ratio for ice that has not yet melted, and pours through a wider straw, and the result is a fountain Diet Coke that qualifies as a separate beverage entirely from the canned. Even a Coke Zero drinker orders it without flinching. It is less a cola than a palate cleanser, and it is the one carve-out my position allows. My friend will read this paragraph as an admission. I maintain it is a footnote.

Two field notes belong in the record. The first is bottle size: Diet Coke in the small glass bottle is a different and better experience than the two-liter, plausibly because it is finished while still cold. The second is that these products travel under aliases. On European flights Coke Zero arrives as Coca-Cola Zero Sugar, in Australia it answers to Coca-Cola No Sugar, and Diet Coke crosses the same borders as Coca-Cola Light, which in much of Europe runs the same aspartame and acesulfame potassium blend as Coke Zero Sugar. In Europe, my friend’s drink is sweetened like my drink. A preference formed at home has to survive translation.

What does preferring Coke Zero say about me? On the surface, that I wanted a Coke, did not want the sugar, and trusted iterative engineering to close the gap. Underneath, something larger: I believe in the engineering and scientific progress of humanity, and I hold it higher than nostalgia. It is the same temperament that would rather work on a new operating system or a new language than hold that the previous ones were perfect. The Diet Coke position, at its most honest, says 1982 was already perfect. Mine says improvement is the point. I believe the 2017 and 2021 reformulations were better, which is to say I believe in version control, for beverages and for everything else.

In the same spirit of disclosure, one more entry in the portfolio. For recovery, after a long run, a long flight, or a long night, nothing yet engineered beats a bottled Mexican Coke: cane sugar, glass, the ceremony of the cap. Blind tests find the cane-sugar difference smaller than the reputation, which changes nothing, because recovery is a ritual before it is a physiology experiment. Expectation is an ingredient, and a cold glass bottle is its best delivery system. I call this an application of my own findings.

The honest verdict is that the scientific evidence for a universal human preference does not exist. What exists is exposure, expectation, and identity, which is why the debate never dies. But if we must score it: on the metric “tastes like Coke,” Coke Zero wins by design. On the metric “tastes like Diet Coke,” Diet Coke remains, as ever, undefeated. And on the metric “recovery,” we both lose to a glass bottle of full-sugar Coke, which contains everything we were avoiding and works anyway.

Nobody prefers a cola. We prefer our cola.