Measured pairing · from 63,696 recipes

What goes with clam?

Measured across 63,696 recipes, the 8 ingredients that co-occur with clam most disproportionately are mussel (×37.6 chance, n=58), squid (×25.8 chance, n=49), enokidake (×23.6 chance, n=20), kelp (×21.6 chance, n=31), scallop (×19.1 chance, n=46), lobster (×17.3 chance, n=20), saffron (×9.6 chance, n=21) and sake (×9.0 chance, n=49). 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. mussel — PPMI 3.63 · ×37.6 chance · n=58 (mussel itself appears in 193 recipes)
  2. squid — PPMI 3.25 · ×25.8 chance · n=49 (squid itself appears in 238 recipes)
  3. enokidake — PPMI 3.16 · ×23.6 chance · n=20 (enokidake itself appears in 106 recipes)
  4. kelp — PPMI 3.07 · ×21.6 chance · n=31 (kelp itself appears in 180 recipes)
  5. scallop — PPMI 2.95 · ×19.1 chance · n=46 (scallop itself appears in 302 recipes)
  6. lobster — PPMI 2.85 · ×17.3 chance · n=20 (lobster itself appears in 145 recipes)
  7. saffron — PPMI 2.26 · ×9.6 chance · n=21 (saffron itself appears in 273 recipes)
  8. sake — PPMI 2.20 · ×9.0 chance · n=49 (sake itself appears in 681 recipes)

Cook the pairing

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

How this was measured

clam appears in 509 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.