Serious Eats: Drinks
Cocktail 101: Families and Formulas
Last week, I discussed cocktail theory, the principles and practices that aid bartenders and home cocktologists in creating new cocktails. I looked at balancing the flavors of a drink, and the role that direction and personal taste play in creating the flavor profile of a particular cocktail.
This week, we'll branch out a little from those basics. We'll look first at families or categories of cocktails, and then we'll look at basic formulas you can use when testing cocktail recipes.
Depending on the cocktail authority you choose to follow, there are a couple of different ways to approach categorizing cocktails. Writing for SE: Drinks last year, Paul Clarke broke the cocktail categories down as follows:
- Ancestrals: The ur-cocktails, usually just a base spirit, sweetener (sugar, simple syrup, or small amounts of liqueur), water, and bitters. Think here of the Old Fashioned, the Improved Gin Cocktail, or the Sazerac.
- Sours: Margaritas, Sidecars, Daiquiris. The category consists of drinks made of liquor, citrus, and a sweetener (either sugar or liqueur).
- Spirit-Forward Cocktails: These drinks take the Ancestrals and add vermouth and sometimes other liqueurs. Manhattan, Martini, Negroni.
- Duos and Trios: Similar to a Spirit-Forward, these take a spirit and mix it with a liqueur; for a trio, you also add cream. I'll say more about these later.
- Champagne Cocktails: French 75, Champagne Cocktail, Seelbach.
- Highballs, Collinses, and Fizzes.
- Juleps and Smashes.
- Hot Drinks: Irish Coffee, Toddies, etc.
- Flips and Nogs: Drinks using whole eggs.
- Pousse Family: A family fallen rightly into disrepute, these are usually layered drinks made of sweet liqueurs. These days, they linger on mostly in layered shots. I'm a snob, so to hell with these.
- Tropical-Style Drinks: Tiki drinks, mostly.
- Punches: Punches, what else?
- Old (and Odd) Birds: The et cetera category, or anything that doesn't fit into other categories—cobblers, bloodies, beer-based drinks.
Another source for cocktail families is Gary Regan's 2003 book, The Joy of Mixology. Regan's categories match up with Clarke's pretty well, but there are some differences, too. I think we can learn a little about cocktail theory by looking at the distinctions, so here are Regan's categories, condensed a bit for space:
- Beer- and Cider-Based Drinks
- Bottled Cocktails: I wouldn't give this its own category because I think it's just a way to batch up large quantities of cocktails.
- Champagne Cocktails
- Cobblers: A class of drinks that has fallen out of currency. It's made from a base spirit (or sometimes a wine or fortified wine) and a sweetener, such as a syrup or a liqueur. You build them in a manner similar to a julep. Fill a glass with crushed ice and stir until a thin film of ice forms on the glass. Garnish generously with fruit.
- Duos and Trios: Clarke borrows this category from Regan's book. For a Duo, you start with a spirit and add a liqueur. An example is the Alaska—gin and yellow Chartreuse—or the Rusty Nail—Scotch and Drambuie. To make a Trio, you add cream or cream-based liqueur. Examples include the White Russian and the Brandy Alexander. What's interesting about this category is how the cream or cream liqueur completely changes the type of drink. An Alaska might be a perfect before-dinner cocktail, whereas a Brandy Alexander is more of a dessert beverage.
- French-Italian Drinks: A class of drinks containing vermouth. Clarke calls these Spirit-Forward, although he includes the Negroni (and similar drinks, such as the Boulevardier) whereas Regan does not. Regan points out an important note: "French" vermouths, also known as dry or white, generally mix better with gin and vodka, whereas "Italian", sweet or red, are better partners with dark spirits such as whiskies and rums.
- Frozen Drinks: One of a couple of categories here that might make you scratch your head. Regan, I must point out, had a decades-long career as a bartender—a career that predates the cocktail renaissance by many years. He emphasizes hospitality in his book, which sometimes means serving frozen Margaritas and Daiquiris, no matter what you personally might think of them.
- Hot Drinks
- Infusions: Again, not sure that limoncello and infused tequilas deserve a separate category, but that's just a quibble.
- Jelly Shots: See also, Frozen Drinks.
- Milanese Drinks: Here he classes any drink that calls for Campari: the Negroni, the Americano. The name "Milanese" refers to Campari's place of origin, but it fails to clarify anything about what kind of drink it is. In addition to quibbling about the name, I'd broaden the category to include any cocktail in which an amaro plays a prominent role. But Gary's onto something important by grouping amari-based cocktails separately; their bitter, herbal notes really do separate them from the Martinis, Mojitos, and Margaritas of other families.
- Muddled Drinks: An oddball category, which includes both the Caipirinha and the Old Fashioned. I prefer Paul's idea of Ancestrals for something like the Old Fashioned.
- Orphans: Again, the things that don't otherwise fit another category.
- Snappers: Savory cocktails. Bloodies, Caesars, etc.
- Tropical Drinks
Thinking in terms of categories and families is a strong framework for understanding mixed drinks. When you think of the Sidecar as a cousin to the Margarita and Whiskey Sour, you begin to understand the similarities of those drinks. But then, there are the differences to think of, too. A Sidecar's citrus component is lemon juice, but the Daiquiri and Margarita use lime. Why? What is it about rum and tequila that pairs better with lime juice?
Formulas: The Great Disagreement
You might not realize there are many disagreements in bartending. I mean, sure, people argue while drinking all the time. But those dust-ups usually involve such complex topics as religion, politics, and sports rivalries. Disagreements in bartending? What's to disagree about?
One source of contention is proportions and formulas for specific drinks. I like to think about this in terms of ratios—1 part X: 2 parts Y: 3 parts Z, for example. David Embury, writer of the 1948 classic The Fine Art of Mixing Drinks, discusses the wide range of proportions you find when studying recipes: "As with other drinks, the proportions vary all over the map, according to the personal whims and individual taste of the author of the recipe."
He cites a recipe that calls for 1 part sweet, 2 sour, 3 strong (with "strong" meaning the base spirit), and another that calls for 6 parts of sour to 1 part of strong. What Embury saw in 1948 is still the case today. Some people swear by the 1 part sweet, 2 sour, 3 strong recipe; some like a 1 part sweet, 2 sour, 4 strong recipe.
Embury's advice? One part sweet, 2 parts sour, and 8 parts strong.
Now, I've tried that recipe. In practice, for a single cocktail, that translates to 2 ounces spirit, to 1/2 ounce citrus juice, to 1/4 ounce simple syrup or liqueur. First problem I have with that recipe is, that's a small cocktail. Second, it's way too dry for my tastes, and it's a little too boozy.
In the aforementioned Joy of Mixology, Gary Regan also discusses the proportion controversy. He discusses Embury's approach and comes to a similar conclusion: it's too sour. Regan proposes the following proportion for making a Sidecar: 3 parts strong, 2 parts sweet, and 1 part sour. Yes, you've read that right; it's a sour in which the sweetness is bumped up higher than the sour.
You can dispute whether that's the right way to make such a drink, but guess what? Regan would agree with you; he presents his preference and then concludes: "The cocktailian bartender should experiment for himself."
Dissent even spilled into the comments section of last week's post. (Disagreements in a comments section? I'm shocked—SHOCKED!—to hear of it.) Someone linked to a David Wondrich recipe for a Whiskey Sour. Wondrich's take on this drink calls for 2 ounces bourbon, 2/3 ounce lemon juice, and 1 teaspoon sugar. That's...even more sour than Embury's formula, which calls for 1/2 ounce citrus.
But bourbon is bourbon, and it's sweet on its own, so I can see this working. I don't think it would work with rye whiskey at all, though, and I'm not sure about its merits with a dry, well-aged cognac.
My point, both last week and today, is this: You cannot take any single formula or recipe and apply it across the board to every cocktail you might make. A specific Manhattan recipe that works beautifully with Rittenhouse Rye and Dolin Rouge vermouth might taste terrible with Maker's Mark and Cinzano.
Regan suggests tasting all your ingredients on their own before you start mixing, and that's great advice. Know the flavor profile of each ingredient and then start thinking about how well it'll mix with others.
Speaking for myself, I don't have a standard formula that I use for, say, all sours. With a Sidecar, my palate finds balance in a recipe that nearly equalizes the sweet and sour. With a Daiquiri, on the other hand, I like 'em dry and tart. A teaspoon of sugar and 3/4 ounce of lime: perfect.