On July 9 the U S government raised the odds of a very strong El Niño to 81 percent. On July 9 every single one of the five I C O coffee indicators printed its highest number of the month. Ten trading days later, all five printed their lowest.

Here is the sentence a roaster is going to hear at every table this month: coffee is up because of El Niño. It is a good sentence. It sounds like weather, which sounds like something nobody can argue with. It also describes July 9, and the calendar has moved on.

Two public documents, both free, both pulled this morning. The first is NOAA’s ENSO Diagnostic Discussion, issued 9 July 2026 by the Climate Prediction Center. The alert status is El Niño Advisory. The second is the International Coffee Organization’s daily indicator table for July — the I-CIP and its four component groups, one line per trading day, published as a PDF that almost nobody in this industry opens.

Put them next to each other and the month has a spine you would not guess from the coverage.

The I C O composite indicator in US cents per pound: the July 9 high, the July 23 low, and where it closed the month on July 30.

That is a fall of 37.54 cents a pound — down 12.1 percent — across ten trading days, followed by a partial recovery. On a single 37,500-pound container, the swing between those two prints is roughly fourteen thousand dollars of green.

Now the part that made me sit up. It is not one indicator that peaked on July 9. It is all five. The composite, Colombian Milds, Other Milds, Brazilian Naturals and Robustas each recorded their highest July value on July 9 — and each recorded their lowest July value on July 23. Five series, two dates, no exceptions.

And the group that fell hardest is the one that should have been most afraid of the forecast. Brazilian Naturals dropped 14.3 percent between those two dates, from 349.86 to 300.00. Robustas fell 7.1 percent. If the market were pricing an El Niño threat to the Brazilian crop, that ranking is backwards.

I am going to be careful here, because this is exactly where a piece like this usually cheats. I cannot tell you NOAA’s release caused the peak. Same-day is not causation, the I-CIP is a lagging daily average of physical differentials rather than a futures print, and a dozen other things moved in July. What I can tell you is that the forecast did not weaken during the fall. It is still an El Niño Advisory. The 81 percent is still 81 percent.

Kept honest. Coincidence of date is not causation, and I have not claimed it is — the July 9 alignment across five series is a real, checkable fact about the table, and the reason for it is not something either document establishes. The I-CIP is not a futures price. It is the ICO’s daily composite indicator of physical prices in US cents per pound, and it can diverge from ICE arabica and robusta contracts, which are what most price commentary actually tracks. The month is incomplete in the source file: the July table’s last populated trading day is July 30, with July 31 blank as published, so “where it closed the month” above means July 30’s 280.50. The container math is illustrative — 37.54 cents on 37,500 lb is about $14,078 — and real contracts carry differentials, quality premiums and timing that this arithmetic ignores. Finally, NOAA’s numbers are as of 9 July; the weekly Niño-3.4 index cited (+1.2°C) was the latest at that writing, not today’s.

Sources, both fetched and read in full today: (1) NOAA / National Weather Service Climate Prediction Center, ENSO Diagnostic Discussion, issued 9 July 2026, ENSO Alert System Status: El Niño Advisory — cpc.ncep.noaa.gov. Verbatim from that discussion: a 97% chance El Niño persists through early spring 2027; an 81% chance of a very strong El Niño during October–December; latest weekly Niño-3.4 index +1.2°C, with Niño-4 at +0.5°C and Niño-1+2 at +2.7°C; next discussion scheduled 13 August 2026. (2) International Coffee Organization, ICO Indicator Prices — July 2026 (I-CIP), in US cents/lb — icocoffee.org/documents/I-CIP.pdf, extracted with pypdf. Every figure above is read directly off that table: I-CIP high 309.80 on 9 July, low 272.26 on 23 July, 280.50 on 30 July, July average 287.34, day-over-day −1.8%; Colombian Milds 415.66 (9 Jul) to 362.89 (23 Jul); Other Milds 390.52 to 339.05; Brazilian Naturals 349.86 to 300.00; Robustas 192.94 to 179.20. All five series record their July high on 9 July and their July low on 23 July; the 31 July row is blank as published.

Two free documents. Nobody put them next to each other.

The first: NOAA's Climate Prediction Center published its monthly ENSO Diagnostic Discussion on 9 July 2026. Status: El Niño Advisory. An 81% chance of a VERY strong El Niño in October–December — one that would rank among the largest since 1950. A 97% chance it persists into early spring 2027.

The second: the ICO's daily indicator price table for July. One line per trading day, published as a PDF, free.

On 9 July, all five ICO indicators printed their highest value of the month. The composite, Colombian Milds, Other Milds, Brazilian Naturals, Robustas. Five series. Same day.

On 23 July, all five printed their lowest.

The composite went 309.80 → 272.26. Down 12.1% in ten trading days. On a 37,500 lb container that's roughly $14,000 of green.

Here's the part I keep turning over. The group that fell hardest was Brazilian Naturals — down 14.3%. Robustas fell 7.1%. If the market were pricing an El Niño threat to Brazil, that ranking is backwards.

Now the discipline, because this is where a post like this usually cheats: I can't tell you NOAA's release caused the peak. Same-day isn't causation. The I-CIP is a physical-price composite, not a futures print, and plenty of other things moved in July.

What I can tell you is that the forecast never weakened while the price fell. It's still an advisory. 81% is still 81%.

So when someone tells you coffee is up because of El Niño, they're describing 9 July. The calendar moved.

Three things I'd actually do with this:

1. Put 13 August on the calendar. That's the next scheduled ENSO discussion. Between now and then no new forecast information exists — so anything that moves your price in the next twelve days isn't the weather outlook changing.

2. Stop watching the composite. It's a weighted blend and it's the least useful of the five for anyone signing a contract. If you buy washed Centrals, your line is Other Milds: 390.52 down to 339.05. If you buy Brazilian naturals, your line moved nearly twice as far as robusta.

3. Read NOAA's second sentence, not just the number. "Even the strongest El Niño events do not lead to the typical impact everywhere." They put the caveat in the same paragraph as the headline. Anyone quoting the 81% without it is quoting half a source.

Both files are free. Neither is behind a login. The whole finding here was reading the dates.

FIRST SOURCE · one verified original-source finding, composed for one reader · this edition: coffee roasters — written for the people who sign green contracts against a number the trade press quotes and almost nobody opens