Measured pairing · from 63,696 recipes

What goes with peach?

Measured across 63,696 recipes, the 8 ingredients that co-occur with peach most disproportionately are blueberry (×5.9 chance, n=24), raspberry (×5.3 chance, n=36), strawberry (×4.2 chance, n=40), orange juice (×3.6 chance, n=55), banana (×2.7 chance, n=24), pineapple (×2.6 chance, n=38), cherry (×2.5 chance, n=23) and pecan (×2.3 chance, n=43). 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. blueberry — PPMI 1.78 · ×5.9 chance · n=24 (blueberry itself appears in 474 recipes)
  2. raspberry — PPMI 1.67 · ×5.3 chance · n=36 (raspberry itself appears in 794 recipes)
  3. strawberry — PPMI 1.44 · ×4.2 chance · n=40 (strawberry itself appears in 1,105 recipes)
  4. orange juice — PPMI 1.28 · ×3.6 chance · n=55 (orange juice itself appears in 1,789 recipes)
  5. banana — PPMI 1.00 · ×2.7 chance · n=24 (banana itself appears in 1,032 recipes)
  6. pineapple — PPMI 0.96 · ×2.6 chance · n=38 (pineapple itself appears in 1,697 recipes)
  7. cherry — PPMI 0.90 · ×2.5 chance · n=23 (cherry itself appears in 1,095 recipes)
  8. pecan — PPMI 0.82 · ×2.3 chance · n=43 (pecan itself appears in 2,215 recipes)

How this was measured

peach appears in 546 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.