About

Selecting a beverage from an extensive menu or a well-stocked pantry involves a combinatorial problem most people solve with habit bias. Studies in decision fatigue (Baumeister, 2011) confirm that repeated low-stakes choices degrade subsequent decision quality. This generator removes that friction. It draws from a curated dataset of 200+ beverages spanning 15 categories - from espresso-based coffee drinks requiring precise water-to-grounds ratios to cocktails governed by 2:1:1 spirit-citrus-sweetener balancing rules. Each result includes an ingredient list to verify feasibility before committing.

Randomness is sourced from the Web Crypto API (crypto.getRandomValues), not Math.random, which uses a PRNG susceptible to pattern repetition in short sequences. Filters for temperature preference, alcohol content, and category narrow the candidate pool before selection, so the output is random within your constraints. Limitation: recipes are canonical approximations. Regional ingredient availability and personal taste thresholds for sweetness (S) or acidity (A) require manual adjustment.

Formulas

The generator uses cryptographically secure uniform random selection. Given a filtered beverage pool of size n, the selected index i is computed as:

i = crypto.getRandomValues(Uint32) mod n

Each beverage has equal selection probability:

P(b_k) = 1n for all k ∈ [0, n − 1]

The spin animation uses cubic ease-out deceleration over duration T (default 2000 ms):

progress(t) = 1 − (1 − tT)³

Where t = elapsed time in milliseconds, T = total animation duration, P(b_k) = probability of selecting beverage k, n = size of filtered candidate pool, progress = normalized animation position from 0 to 1.

Reference Data

Category	Example Beverages	Typical Temp.	Alcohol	Complexity Range
Espresso Coffee	Latte, Cappuccino, Americano, Macchiato	Hot / Iced	No	1 - 3
Brewed Coffee	French Press, Pour Over, Cold Brew, Turkish	Hot / Cold	No	2 - 4
Black Tea	Earl Grey, English Breakfast, Chai, Darjeeling	Hot	No	1 - 2
Green & Herbal Tea	Matcha, Sencha, Chamomile, Peppermint	Hot / Cold	No	1 - 3
Cocktails	Margarita, Old Fashioned, Mojito, Negroni	Cold	Yes	2 - 5
Mocktails	Virgin Mojito, Shirley Temple, Roy Rogers	Cold	No	2 - 4
Smoothies	Berry Blast, Tropical Mango, Green Detox	Cold	No	1 - 3
Fresh Juice	Orange, Apple-Ginger, Carrot-Beet, Watermelon	Cold	No	1 - 2
Lemonade	Classic, Lavender, Strawberry, Arnold Palmer	Cold	No	1 - 2
Soda & Fizzy	Ginger Beer, Root Beer, Italian Soda, Tonic	Cold	No	1
Beer	IPA, Stout, Wheat, Pilsner, Porter, Lager	Cold	Yes	1
Wine	Cabernet, Chardonnay, Rosé, Prosecco, Sangria	Cold	Yes	1 - 3
Spirits (Neat/Rocks)	Whiskey, Rum, Tequila, Gin, Vodka, Brandy	Room / Cold	Yes	1
Hot Chocolate	Classic, Mexican Spiced, White, Peppermint	Hot	No	1 - 3
Milkshakes	Vanilla, Chocolate, Strawberry, Oreo, Peanut Butter	Cold	No	1 - 2
Energy Drinks	Citrus Blast, Berry Rush, Tropical Surge	Cold	No	1
Infused Water	Cucumber-Mint, Lemon-Ginger, Berry, Citrus	Cold	No	1

Frequently Asked Questions

The generator uses the Web Crypto API (crypto.getRandomValues) which produces cryptographically secure pseudo-random numbers from the operating system's entropy pool. This avoids the PRNG bias of Math.random(). The modulo operation against the pool size n ensures uniform distribution with negligible bias when n is much smaller than 2^32 (≈ 4.29 billion), which is always the case for our dataset of ~200 beverages.

Yes. True randomness does not exclude repetition. Each spin is an independent event with probability P(b_k) = 1/n. If you want to avoid repeats, enable the "No Repeats" toggle, which removes previously generated beverages from the candidate pool until all options in the filtered set are exhausted, then resets.

Filters reduce the candidate pool size n before random selection occurs. For example, selecting only "Cold" and "Non-Alcoholic" from a full set of 200 beverages might yield n = 85 candidates. Each candidate then has probability 1/85 ≈ 1.18%. The generator recalculates the pool on every spin, so changing a filter immediately affects the next result.

Difficulty reflects preparation complexity: 1 = ready to drink or single-step (open a can, pour water), 2 = basic mixing (2-3 ingredients, no special equipment), 3 = moderate technique (steaming milk, muddling herbs), 4 = advanced (multiple steps, precise timing, specialized tools like cocktail shaker), 5 = expert (layered drinks, infusions, sous vide preparation). These ratings follow bartender guild training level classifications.

The ingredient list is displayed transparently so you can assess compatibility. Beverages containing dairy (milk, cream, butter) are identifiable by scanning the ingredients. The generator does not currently filter by allergen, so manual verification against your specific dietary requirements is necessary. Common substitutions: oat milk for dairy in lattes, agave for honey in cocktails, coconut cream for heavy cream in smoothies.

Classic cocktail architecture follows the 2:1:1 ratio (spirit : citrus : sweetener) established by Jerry Thomas in 1862. Deviation beyond ±15% creates perceptible imbalance. A Margarita at 2:1:1 (tequila : lime : triple sec) yields ~18% ABV in the glass. Shifting to 3:1:1 pushes to ~22% ABV with noticeable burn. The recipes in this generator use canonical ratios from IBA (International Bartenders Association) official specifications.