Last night we threw a surprise birthday party for my dad and decided to conduct a little bit of an experiment. The guest of honor is somewhat of an ice cream connoisseur and has long debated the merits of Bi-Rite versus Mitchell's. We decided to settle, once-and-for-all, which reigns supreme by having a head-to-head taste off. Naturally, we dubbed it The 2009 Alan R. Fedman Ice Cream Taste Off.
It didn't take long for the participants in this taste off to voice their allegiances, which brought the divisive nature of the debate into sharp focus. One participant declared Mitchells, "the people's ice cream," perhaps in reference to its more affordable price tag or what some perceive to be its laid-back counterpoint to the alleged snootiness of Bi-Rite. Others defended Bi-Rite, lauding its natural ingredients and widespread popularity. The room was ripe with conflict, and it was time to dish out the ice cream.
The methodology:
We started by analyzing the flavors offered by both establishments, looking for an overlap in flavors. We discovered ten:
Vanilla
Chocolate
Minty
Cinnamon Snap
Banana
Pumpkin
Tea
Ginger
Sorbet
Cookies and Cream
In the end, we offered four for the taste-off: ricanelas/cinnamon snap, pumpkin, cookies and cream, and mint chocolate chip.
One obvious flaw with this experiment is that these aren't even close to either establishment's best flavors, nor do they account for the fun, off-the-wall, creative flavors that both places thrive in. But they do provide a sample that allows us to objectively compare.
We had eleven participants. Each was offered two spoonfuls of each sample - labeled either A or B. We took care to ensure that each flavor was served so that it was virtually impossible to discern its maker (with one flawed exception that you will see below). We had four rounds (one for each flavor) and each participant filled in a score card by circling their preference for the round: either A or B. Participants were prohibited from voicing their preference throughout the entire process so as not to influence the opinions of others. They were offered either water or beer to cleanse their palate between rounds.
Whichever ice cream received more votes for each round (out of eleven), received one point. Whichever maker received more points overall was the victor.
The results are as follows:
Round 1 (Cinnamon)
Bi-Rite: 10
Mitchells: 1
Point Bi-Rite.
This, notably, was the most decisive preference in the entire experiment. Bi-Rite's Ricanelas is damn good, and far superior to Mitchells.
Round 2 (Thin Mint)
Bi-Rite: 4
Mitchells: 7
Point Mitchells.
It should be noted that there is a potentially confounding variable with this round in that the color of both selections is different (one green, one white), which could have potentially revealed its maker to the trained ice-cream-ophile. That is, I think a few people (some of whom had strong allegiances to Mitchells) knew which they were eating in this round.
Round 3 (Pumkin)
Bi-Rite: 5
Mitchells: 6
Point Mitchells.
As the results show, this was hands down the closest decision of the entire taste-off. Both were very similar, and both were very delicious.
Round 4 (Cookies and Cream)
Bi-Rite: 4
Mitchell's: 7
Point, Set, Match, Mitchells.
Here, again, it was a close call, but Mitchells decisively prevails once more and thus solidifies its place as the superior ice cream, 3 to 1.
One especially sharp participant pointed out that, interestingly, Bi-rite has the majority of points in the overall voting, but loses to the electoral system of the experiment.
A couple of points of clarification. I acknowledge that we ran the risk of ending up with a tie having offered four flavors, but it fortunately didn't become a problem. Perhaps the most contentious part of the methodology here was that head-to-head nature of this competition. Some participants voiced a frustration with the experiment as overly simplified. They would have preferred a more nuanced criteria based on texture, taste, creativity, and so forth. While I agree that it should be done in the future, it was simply too much number crunching for one night.
No comments:
Post a Comment