Measured pairing · from 63,696 recipes

What goes with garlic?

Measured across 63,696 recipes, the 8 ingredients that co-occur with garlic most disproportionately are celery oil (×3.2 chance, n=1008), thai pepper (×3.1 chance, n=133), chinese cabbage (×2.8 chance, n=153), oregano (×2.5 chance, n=2627), mussel (×2.4 chance, n=144), dried leaf oregano (×2.4 chance, n=135), white onion (×2.3 chance, n=75) and black bean (×2.3 chance, n=444). The score behind the ranking is positive pointwise mutual information — how many times more often a pair shares a recipe than random combination would predict — and n is the number of recipes containing both ingredients. Nothing is published below a floor of 20 recipes per pair and 100 per ingredient. Co-occurrence records what recipe authors actually do — it conflates deliciousness with popularity and tradition — so read it as ethnography, not judgment.

The measured partners — PPMI · ×chance · n

  1. celery oil — PPMI 1.16 · ×3.2 chance · n=1008 (celery oil itself appears in 1,008 recipes)
  2. thai pepper — PPMI 1.13 · ×3.1 chance · n=133 (thai pepper itself appears in 137 recipes)
  3. chinese cabbage — PPMI 1.05 · ×2.8 chance · n=153 (chinese cabbage itself appears in 172 recipes)
  4. oregano — PPMI 0.90 · ×2.5 chance · n=2627 (oregano itself appears in 3,429 recipes)
  5. mussel — PPMI 0.87 · ×2.4 chance · n=144 (mussel itself appears in 193 recipes)
  6. dried leaf oregano — PPMI 0.86 · ×2.4 chance · n=135 (dried leaf oregano itself appears in 183 recipes)
  7. white onion — PPMI 0.85 · ×2.3 chance · n=75 (white onion itself appears in 103 recipes)
  8. black bean — PPMI 0.84 · ×2.3 chance · n=444 (black bean itself appears in 611 recipes)

Cook the pairing

Dishes in Palate's live corpus that use garlic with one of its measured partners:

How this was measured

garlic appears in 19,906 of the 63,696 recipes counted (5,296 parsed + 1,081 hand-curated + 57,319 from the dataset published with Ahn et al., 2011). PPMI = max(0, ln(count(A,B)·N / (count(A)·count(B)))), all counts over the same N; lift = e^PMI, which reads directly as "×chance." Staples (salt, sugar, flour, water) are excluded because sources under-record them, and pairs where one name contains the other are excluded as near-duplicates.

Read the numbers with their limits: the corpus is Western-skewed (90% of it is the largely North American and European recipe dataset published with Ahn et al.'s 2011 paper), so the absence of a pairing here proves nothing except who wrote the recipes down. PPMI favors the rare end even above the floor — a niche pair at n=22 can outrank a workhorse pair seen thousands of times, because common ingredients' huge counts divide the score down — so read PPMI and n together.

The full reference — 150 ingredients, top 8 partners each — lives at What actually goes with X.

Curious how these flavors fit your palate — take the short taste test →

Every number above is counted from recipes — pointwise mutual information over 63,696 of them — not chef interviews, model guesses, or reviews.