taking the first observations…
CLEAR
0
0% of stations
BREEZE
0
0% of stations
GALE
0
0% of stations
Barographmean reading · 24h · UTC
the pen has not inked a full hour yet
hourly mean of every reading taken · tendency is the change over three hours, as a station reports it
Ithe office

A forecast office, not a scanner.

A scanner prints a number and walks away. An office publishes a forecast, waits, and then goes back to mark it — in public, whether the mark is flattering or not. That second half is the entire difference, and it is the half almost nobody builds.

Iso- is Greek for equal. Isobars join points of equal pressure; isotherms, equal temperature. ISOU draws its own — lines joining points of equal forecast across every live pair on Solana.

0
stations on the chart
12
instruments per reading
6h
until every forecast is marked
iso-lines · points of equal forecast
IIthe instruments

Twelve readings, and a rule for what happens when one is blind.

Two come from the chain itself — whether the deployer can still mint supply or freeze your wallet. The rest read the market's shape. Each carries a published weight and hard bounds that verification may never push it past.

the office is not reporting its instrument list
The rule that matters most

Sources omit fields. DexScreener leaves liquidity out entirely on very fresh pairs. When that happens the instrument does not fire and is reported unread — it is never quietly scored as zero, because that would turn silence into evidence. Every station report shows its coverage, and the chart counts the stations it could not place rather than placing them at the origin.

IIIthe bands

Three bands, named the way a forecast should be.

the field right now — 0 stations
CLEAR≥ 7.5
0
0% of stations

The instruments found little wrong. Not a recommendation — an absence of findings.

BREEZE4.5 – 7.5
0
0% of stations

Mixed. Something is unsettled, and the reading says which instrument saw it.

GALE< 4.5
0
0% of stations

The office is warning you. Usually authority still live, or depth that cannot absorb an exit.

IVverification

The loop closes. That is the whole product.

OBSERVE01
Every 45 seconds

Fresh pairs are pulled and stations whose instruments may have moved are read again.

FORECAST02
0 – 10, with a band

The reading starts at 5.0; each instrument that registers moves it by a published weight.

VERIFY03
Six hours later

The pair is read again and the forecast is marked RUG, FADED or HELD against what happened.

REVISE04
A new model run

Instruments that called it earn weight; instruments that cried wolf lose it. Bounded, logged.

Vthe record

A forecast nobody marks is a horoscope.

So the office keeps a record of how wrong it has been, and publishes it whether or not it flatters the office. It is the first page to read, not the last.

forecasts verified
0

Each one read again six hours after it was issued, and marked against what the pair actually did.

skill over the field

Under twenty verified forecasts the office states no skill figure at all. A number drawn from a handful of marks says more about luck than about the model.

what we will not do
  • — restate a figure once published
  • — score a missing reading as zero
  • — show a number the sample cannot support
  • — call a reading a recommendation