WHAT’S THE BEST COFFEE IN THE WORLD?

The reason why is simple. There is no single flavor profile that everybody is going to think is the best coffee they’ve ever drunk. What would that coffee taste like? The millions (billions?) of coffee drinkers on the planet surely will all have something a bit different to say about it. How should it be processed? At what roast level should it be? Should it have lots of complexity and nuance or just really taste like coffee?

There is no goldilocks coffee. There is no way to define quality in a way that suits every person, every time. Rather, there are many “best” coffees in the world. Preference is subjective and this means diversity is something to celebrate. Thus, the only sufficient answer to this question is “whatever coffee you like the most.”

The only way to talk about quality in an absolute sense is if it is first defined. This means laying out all the organoleptic characteristics and assigning them the desired intensity. One way to do this is to establish a scale of 1 to 10 where the range represents “not present” to “so present no other coffee can have more of this”. Then, each characteristic and descriptor is assigned a score: acidity = 3, body = 8, floral = 6…. Once a definition is established, there can be a standard by which to measure and thus a best coffee or a winner in a competition. Until then, “best” is always a moving target.

After a definition is created, we can do cool things like taste any coffee in the world and say whether it is good or not good relative to that standard. All we have to do is compare its intensity values to the standard (this is trickier than it sounds, but not impossible). To make communication simpler, we can make arbitrary categories like horrible, bad, poor, acceptable, good, great, and spectacular that all describe how far away a particular coffee is from the standard. 

We can get really geeky and graph the probability of randomly picking a coffee of a particular quality from a set of coffees. To do this, we first need a set, or a population, of samples from which to make our selections. Let’s say our population is composed of all the coffee farms in a country and a randomly picked sample is a single farm.

Our graph will have two axes. The x-axis (horizontal) will be “quality” with movement towards the right getting us closer to the standard we’ve defined and the far left being a level of quality as far away from that standard as possible. The y-axis (vertical) will be “number of farms.” If we plot our population on the graph, it will look like a bell curve where the tail end on the left represents horrible, the tail end on the right represents spectacular, and the other categories fall into place in the middle.

The areas under the curve that our categories represent are not chosen arbitrarily.