Measured pairing · from 63,696 recipes

What goes with oat?

Measured across 63,696 recipes, the 8 ingredients that co-occur with oat most disproportionately are grain wheat flour (×9.2 chance, n=134), peanut butter (×6.1 chance, n=128), raisin (×6.0 chance, n=235), date (×5.0 chance, n=38), coconut (×4.6 chance, n=170), corn flake (×4.3 chance, n=20), walnut (×4.0 chance, n=224) and cranberry (×3.7 chance, n=68). 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. grain wheat flour — PPMI 2.22 · ×9.2 chance · n=134 (grain wheat flour itself appears in 733 recipes)
  2. peanut butter — PPMI 1.82 · ×6.1 chance · n=128 (peanut butter itself appears in 1,046 recipes)
  3. raisin — PPMI 1.79 · ×6.0 chance · n=235 (raisin itself appears in 1,963 recipes)
  4. date — PPMI 1.61 · ×5.0 chance · n=38 (date itself appears in 382 recipes)
  5. coconut — PPMI 1.52 · ×4.6 chance · n=170 (coconut itself appears in 1,858 recipes)
  6. corn flake — PPMI 1.46 · ×4.3 chance · n=20 (corn flake itself appears in 233 recipes)
  7. walnut — PPMI 1.39 · ×4.0 chance · n=224 (walnut itself appears in 2,816 recipes)
  8. cranberry — PPMI 1.31 · ×3.7 chance · n=68 (cranberry itself appears in 919 recipes)

How this was measured

oat appears in 1,268 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.