No difference

The "solunar theory" anglers swear by

Fishing has its own old moon lore. The positions of the moon and sun are said to govern when fish are active, and the bite is supposedly best around the full and new moon — the spring tides. This is "solunar theory," proposed by the American John Alden Knight in the 1930s and built into fishing-forecast apps ever since. Plenty of anglers still cast their lines because "the tide is big today."

If fish were really caught more then, that fish would travel from port to market and across the auction floor. So if the full moon moved the fish, the volume reaching the market should rise around it. We test that against ~22 years of daily wholesale volume at Tokyo's central wholesale market — Japan's largest fish market.

How the verdict is computed

• Data is the daily wholesale seafood volume from the Tokyo Metropolitan Central Wholesale Market's daily report (2004 onward; every trading day's Shift-JIS CSV aggregated). Beyond the total we split out fresh fish, where a fishing effect would show first, and frozen, which the moon cannot possibly move — a built-in control
• A fish market has a strong weekday rhythm (closed Sundays, some Wednesdays) and seasonality (the season's catch, year-end demand). So we use the "same weekday × same month" average as the expected value and compare actual ÷ expected — the same adjustment as the births and accidents topics
• We mark "full moon days" (instant ±24h) and "new moon days" from the lunar phase at noon Japan time, then compare each group's mean index against normal (100)

See the methodology for the full criteria.

Reading the data with care

Market arrivals are a proxy for supply, not the catch itself. Landed fish take roughly a day to reach the floor, and deep-sea frozen tuna flows year-round, almost detached from moon or season. So this data reflects the rhythm of the coastal fresh catch to a degree, but it is not a complete record of how much was caught. Even so, if the full moon swayed fishing meaningfully, its shadow should fall across thousands of trading days.

The spring tide: a real clue

"The bite is better on the spring tide" rests on a real clue. The spring tide is the moon's own work — on this site too, tides came out clearly "higher" (see tides). But market volume is also pushed hard by the season, the year-end table, and the market's opening rhythm. The moon's part may be getting lost among those bigger waves before it reaches the market's numbers.

What we can't see yet

This result doesn't mean "the moon has nothing to do with it." Here is what the current method can't see:

• How fish bite: an angler's catch and the volume reaching market are different things. A great day on the water doesn't translate directly into market volume
• Species: if only certain coastal fish respond to the spring tide, they're hard to see inside the total of all fish
• Timing: spring tides peak one to two days after the full moon, and fish take about another day to reach the market. Looking only at full-moon days may miss the peak

What we want to check next

• Split fresh fish by species, such as horse mackerel and squid
• Shift the window two to three days after the full moon
• Check catch records from fishing boats

Perhaps anglers' sense that "they bite on the spring tide" holds a link between the moon and fish. We'd like to find it in the data one day.

Sources

• Tokyo Metropolitan Central Wholesale Market daily report (daily wholesale seafood volume, all markets; source: Tokyo Metropolitan Central Wholesale Market)
• Lunar age and the instant of each phase are computed in-house from the algorithms in Jean Meeus, Astronomical Algorithms (UTC basis)