Measured pairing · from 63,696 recipes

What goes with celery oil?

Measured across 63,696 recipes, the 8 ingredients that co-occur with celery oil most disproportionately are corn (×12.8 chance, n=1008), tamarind (×12.7 chance, n=338), tabasco pepper (×7.9 chance, n=122), vinegar (×7.7 chance, n=1008), cider (×7.1 chance, n=128), horseradish (×6.7 chance, n=45), tomato (×5.8 chance, n=1008) and mustard (×5.5 chance, n=368). 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. corn — PPMI 2.55 · ×12.8 chance · n=1008 (corn itself appears in 4,964 recipes)
  2. tamarind — PPMI 2.54 · ×12.7 chance · n=338 (tamarind itself appears in 1,682 recipes)
  3. tabasco pepper — PPMI 2.07 · ×7.9 chance · n=122 (tabasco pepper itself appears in 977 recipes)
  4. vinegar — PPMI 2.04 · ×7.7 chance · n=1008 (vinegar itself appears in 8,238 recipes)
  5. cider — PPMI 1.96 · ×7.1 chance · n=128 (cider itself appears in 1,135 recipes)
  6. horseradish — PPMI 1.90 · ×6.7 chance · n=45 (horseradish itself appears in 425 recipes)
  7. tomato — PPMI 1.75 · ×5.8 chance · n=1008 (tomato itself appears in 11,069 recipes)
  8. mustard — PPMI 1.70 · ×5.5 chance · n=368 (mustard itself appears in 4,243 recipes)

How this was measured

celery oil appears in 1,008 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.