Measured pairing · from 63,696 recipes

What goes with onion?

Measured across 63,696 recipes, the 8 ingredients that co-occur with onion most disproportionately are tamarind (×3.0 chance, n=1633), kidney bean (×2.6 chance, n=441), green bell pepper (×2.6 chance, n=2283), red kidney bean (×2.5 chance, n=82), lentil (×2.5 chance, n=214), tomato juice (×2.4 chance, n=171), green pepper (×2.4 chance, n=247) and bay (×2.4 chance, n=1219). 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. tamarind — PPMI 1.09 · ×3.0 chance · n=1633 (tamarind itself appears in 1,682 recipes)
  2. kidney bean — PPMI 0.96 · ×2.6 chance · n=441 (kidney bean itself appears in 518 recipes)
  3. green bell pepper — PPMI 0.94 · ×2.6 chance · n=2283 (green bell pepper itself appears in 2,733 recipes)
  4. red kidney bean — PPMI 0.93 · ×2.5 chance · n=82 (red kidney bean itself appears in 100 recipes)
  5. lentil — PPMI 0.90 · ×2.5 chance · n=214 (lentil itself appears in 267 recipes)
  6. tomato juice — PPMI 0.89 · ×2.4 chance · n=171 (tomato juice itself appears in 217 recipes)
  7. green pepper — PPMI 0.88 · ×2.4 chance · n=247 (green pepper itself appears in 316 recipes)
  8. bay — PPMI 0.87 · ×2.4 chance · n=1219 (bay itself appears in 1,568 recipes)

Cook the pairing

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

How this was measured

onion appears in 20,685 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.