Coffee Review - The World's Leading Coffee Guide

Kenneth Davids' Response to Mark Prince's Column in Coffee Geek, September 2007

Earlier this month (September 2007) Mark Prince, editor of the lively and useful website generously praised my work as cupper and coffee writer. (Thank you Mark.) He then launched into an unsubstantiated and rather rambling attack on my credibility as a judge of espresso, provoked apparently by my recent review article Italy Seen from America: Nine (Genuine) Italian and Three American Espressos. (Visit to view Mark's column).

For Mark the fact that I rated the mainstream Italian blend Segafredo Massimo, for example, higher than the American Black Cat blend from Intelligentsia indicated that something had to be wrong -- with me, with my espresso machine, or with my protocol. Along the way he also raised a couple of interesting and genuine issues about how to best review espresso coffees for consumers. I get to these issues later in this piece.

Just as Mark prefaced his attack with praise for my other work and books, I preface this response with praise for Mark and his work and his community-building website. I do think he should check out rumors before assuming they are truth, however, and perhaps engage with a bit more depth in his subject before firing his cannons.

First, here is a sort of executive summary of my response. Following this summary is more detail on my brewing protocol and history at Coffee Review.

Mark's Accusation
My espresso brewing protocol is suspect.

I use a La Marzocco semi-automatic Linea 1-group with PID (Proportional-Integral-Derivative) control of brewing temperature, and use the same protocol Mark proposes and used elsewhere in the industry. Aside from a couple of experiments some years ago, grinding and dosing and extraction have always been controlled manually. See below for details. The protocol is outlined on the site and embedded in some past articles (Readers' Choice Espressos, October 2006, for example). I did not embed the protocol in the current article, however, figuring that to do so would be redundant. Lesson: There is no such thing as redundancy on the Internet.

I can't pull good shots.

I have been pulling shots off and on for forty years, and furthermore have been honing my skills recently on my La Marzocco. I definitely lack the grace necessary to compete in a barista contest, but I also feel that I can do as well as anyone in pulling the best shot possible from a stranger's espresso blend.

Mark's Assumption
My article represented a comprehensive Italy-vs-USA test of espresso blends.

I have the greatest respect for American boutique specialty espressos and have consistently given many of them higher ratings than I gave any of the Italian espressos in the article that set Mark off (for example, see Better than Ever: Boutique Espressos). Nor did I present my article as a comprehensive Italy-vs-USA (or North America) test. As I tried to make it clear in title and introduction, this set of reviews, like all of my reviews, is a momentary snapshot of a small slice of the industry. All I was doing was testing Italian espressos currently available to consumers on the Internet against a very small basket of American espressos from medium-to-larger roasting companies. As I pointed out in the introduction, I did not source coffees from boutique American roasters because neither I nor American consumers have ready access to coffees from similar boutique roasters based in Italy.

I should have altered brewing temperature to suit the characteristics (particularly the degree of roast) of the blends being tested.

Depends on whether the reviews are aimed at consumers, who (in almost all cases) can't alter brewing temperatures on their home machines, or food service customers of roasting companies, who can set temperatures to match their suppliers' blends. We aim our reviews at consumers, not retailers, so I have kept the brewing temperature on the Marzocco set at 200°F, which in my reading of the debate on temperature is a good compromise setting.

On the other hand, Mark Prince points out that some aficionado espresso lovers are able to tweak brewing temperature on their home machines using a variety of improvised strategies.

Frankly, our reviews are aimed at all users of home espresso machines, not just espresso hobbyists. Nevertheless, I've decided that it can't hurt to roll over on this issue and invite companies who submit espressos for review at Coffee Review to specify the temperature at which they would like to have their espressos reviewed. If they have no preference or do not respond, I will maintain my current default brewing temperature of 200°F. I'll report the brewing temperature in the "Notes" section of the review.

I failed to take into account that Italian espressos are released with a very long, other-worldly shelf-life assumption.

I simply reviewed what came out of the bag because I assume this is what a consumer is likely to taste. Only two of the Italian espressos I tested showed signs of staling, and I noted those in my review. It is quite likely that even the best packaged coffees stale faster out of the bag than do fresh bulk coffees. But I am assuming that a good, coffee-involved consumer is not likely to keep coffee around for longer than two weeks anyhow. I probably should have mentioned this issue in my introduction. But certainly I was not about to artificially deduct points from a rating because possibly or hypothetically the contents might stale quicker over the course of a couple of weeks.

The Importance of Blind Tasting

Mark indirectly accuses me of being too engaged in the protocols of cupping to make a proper judge of espresso. What I think he is implying here is that espresso should only be judged in the full context of the visual and performance aspects of its production. In other words, it appears that from Mark's point of view that the crema, the pour, and (in the case of barista contests) the entire magnificent theater of drink production should be taken into account when evaluating espresso.

I have tremendous admiration for the stirring spectacle of barista contests, and certainly I admit that a shot of espresso with a coarse or button-holed crema is a depressing sight (though I have never witnessed such shabby crema in any of the espressos I have reviewed over the last ten years).

Nevertheless, I am a thorough-going adherent of blind testing of the non-visual, gustatory properties of beverages. If there is a difference in emphasis here between the kind of more performative evaluation Mark proposes and the kind of blind, purely mouth-and-nose based sensory testing I carry out, then vive la difference. Together we can reach a better understanding of a beverage we both love. But I certainly see no reason for me to abandon my protocol, which emphasizes the need to repress any visual cues that might trigger prior assumptions or preferences.

I have found over the fifteen years or so that I have been reviewing coffees that the most destructive thing one can do to solid, impartial reviewing is to make any assumptions whatsoever about what one is tasting that might interfere with the pure, focused activity of registering smell, taste and mouthfeel.

I never start a cupping or a tasting with the idea that I am going to "prove" anything. I chose a general category of coffees to examine, and then listen to the coffees with as great and detached attention as I can muster. In other words, I taste the coffees, write the reviews, and then, finally, I write the introduction based on whatever potentially interesting themes emerge from experiencing the coffees.

In the case of the contested article, I did not set out with any prejudices whatsoever in favor of Italian mainstream espressos. In fact, I expected them to be half stale from ancient packaging and a bit flat owing to the presence of too many Brazils and robustas. If anything, I was prejudiced in favor of the Intelligentsia Black Cat blend because I admire Intelligentsia as much as any coffee company for its no-compromise pursuit of quality, and always go out of my way to recommend it when I speak to journalists.

But the tests came out the way they did, I was struck by some interesting similarities in approach among the Italian espressos that I thought perhaps we could learn from, and explored that aspect of the experience in my introduction.

Some Personal Detail

My Espresso Skills
To repeat, I currently I use a one-group La Marzocco Linea semi-automatic fitted with PID temperature control with matching grinder. All grinding and dosing is manually controlled. I have been pulling shots off and on for 40 years — I owned a café in Berkeley in the 70's when I first started in coffee and often worked behind the counter on a Gaggia lever machine. But over the last two years I have become especially proficient in the use of the La Marzocco and feel that I pull as good a shot as anyone.

Coffee Review Espresso Protocol
Aside from a couple of experiments some years ago I currently follow and have always followed the exact protocol Mark describes in his column — double portafilter, leveling technique, which nets 14 — 15 grams per double, pull shots into two glasses, a shot glass and a small glass to which I add two parts of hot (not frothed) milk for the "with milk" part of the evaluation. Tamped at the usual 30 to 40 pounds of pressure with a light polish applied to the dose.

I aim for as close as possible to 1 ounce of liquid (never more) once the crema has settled.

Lately (as I did for this review) I have been pulling at least three shots at different extraction times. I base the review on the shot that seems to show the coffee at its best. These shots tend to run from 18 seconds after the first drop (23 seconds after initiating the extraction on the La Marzocco) to a high of about 40 seconds after the first drop (45 after initiating the shot). Usually the shot that comes in at around 25/30 is the one on which I base the review.

Many of my early reviews were based on shots produced by other experienced baristas, but always on La Marzocco equipment using the standard protocol Mark references. The reasoning behind employing other baristas was to eliminate the bias that comes from watching the pour and to permit full, immediate focus on the freshly pulled shot. I also tried to have fellow tasters during those early reviews. I was always aware, as I am now, of the challenges of reviewing a beverage as subtle and tactile as espresso, so I felt more comfortable if I had a fellow experienced taster to confirm my judgments.

I admit that for one or two reviews I experimented with a home super automatic machine. My reasoning at the time was that I was writing for consumers who hardly would have access to professional equipment, and a home super automatic provided a sort of level playing field for espressos. As a matter of fact, I think it did — the best espressos stood up fine in that brewing environment.

However, I eventually concluded that I had to support the refinements going on in the industry, so I went back to pulling shots on the La Marzocco and a conventional grinder. That was at least three years ago, with most of the reviews since then based on shots on the Marzocco in my lab but one review based on shots pulled at the Specialty Coffee Association of America lab in Long Beach, where I was joined by Ted Lingle as co-taster. Ultimately, I found that I could do a more deliberate, better evaluation pulling the shots myself in my lab because I was able to take my time and keep recalibrating the grinder as often as it took to get the right shot. Third persons doing the shots often rushed ("it's good enough") because they had two tasters sitting around fiddling with their pencils waiting for the next round.

The irony of the timing of Mark Prince's attack is that the contested review of Italian and three American espressos was carried out with particular care because I had relatively few coffees to test and could take the time to be even freer than usual to obsess over the shots. The shots pulled at Bellissimo during a workshop that Mark mentions in his initial attack had no bearing whatsoever on the review results, by the way. I had finished the story before I gave the workshop. But I will say that the participants in the workshop were even more enthusiastic than I was about the top-rated Segafredo Massimo (they had no idea whose espresso they were tasting at the time) and their response generally mirrored my assessments.