Measured pairing · from 63,696 recipes
What goes with fish?
Measured across 63,696 recipes, the 8 ingredients that co-occur with fish most disproportionately are chinese cabbage (×19.6 chance, n=112), kelp (×14.4 chance, n=86), lemongrass (×13.2 chance, n=116), thai pepper (×10.5 chance, n=48), enokidake (×10.2 chance, n=36), seaweed (×8.4 chance, n=60), radish (×8.0 chance, n=147) and squid (×6.3 chance, n=50). 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
- chinese cabbage — PPMI 2.98 · ×19.6 chance · n=112 (chinese cabbage itself appears in 172 recipes)
- kelp — PPMI 2.67 · ×14.4 chance · n=86 (kelp itself appears in 180 recipes)
- lemongrass — PPMI 2.58 · ×13.2 chance · n=116 (lemongrass itself appears in 264 recipes)
- thai pepper — PPMI 2.35 · ×10.5 chance · n=48 (thai pepper itself appears in 137 recipes)
- enokidake — PPMI 2.32 · ×10.2 chance · n=36 (enokidake itself appears in 106 recipes)
- seaweed — PPMI 2.13 · ×8.4 chance · n=60 (seaweed itself appears in 215 recipes)
- radish — PPMI 2.08 · ×8.0 chance · n=147 (radish itself appears in 555 recipes)
- squid — PPMI 1.84 · ×6.3 chance · n=50 (squid itself appears in 238 recipes)
Cook the pairing
Dishes in Palate's live corpus that use fish with one of its measured partners:
- Vietnamese Beef Stew — with lemongrass, radish
- Baechu Kimchi (Whole Napa Cabbage Kimchi) — with chinese cabbage, radish
- Kimchi (김치) — with radish
- Tom Yum Goong — with lemongrass
- Tom Kha Gai — with lemongrass
- Satay (Moo Satay) — with lemongrass
How this was measured
fish appears in 2,118 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.