Apex Predictive
Dozens of faint cool traces of light streaming from the left, braiding together and converging into one bright golden line that continues alone to the right.

Apex Predictive

Every future, argued.
One future, published.

A circle of independent expert models works every question. Most weeks they disagree, and we publish nothing. When they converge with high confidence, the prediction goes on the board — and on Friday, we grade it in public.

Monday
The week's board is published: ten predictions the circle agreed on.
Friday
Every prediction is judged against what actually happened. Hits and misses, in the open.
The gate
No consensus, no publication. Silence is the engine working correctly.

The premise

One forecaster is an opinion.
A converged circle is a signal.

Ask one model about the future and you get a guess with good grammar. Apex Predictive asks many — independent models with different training, different methods and different failure modes — and lets them work the same question separately, each with live data, none seeing another's answer.

Then comes the only part that matters: the gate. If the circle scatters, the question dies quietly. Only when the experts land on essentially the same outcome, at high confidence, does a prediction exist at all. Disagreement is information, and the information is do not publish.

Stock closes, weather, elections, launches, box office, court calendars — anything with a verifiable outcome is fair game, because the whole system is built around being checkable.

The board Mon 9:00 AM CT → Fri 4:00 PM CT

Ten a week.
Judged in public.

This is the shape of the product — a sample board, with illustrative statements. The live board arrives at launch, and once it does, every week's results stay on the record forever.

The Week's Board — Sample Illustrative · format preview, not live predictions
  1. 01

    The S&P 500 closes Friday above its Monday open.

    Markets · resolves Fri 3:00 PM CT

    11/12
    Pending
  2. 02

    Measurable rain falls at DFW Airport before Friday noon.

    Weather · resolves Fri 12:00 PM CT

    12/12
    Pending
  3. 03

    Initial jobless claims print below the consensus estimate on Thursday.

    Economy · resolves Thu 7:30 AM CT

    11/12
    Pending
  4. 04

    The scheduled coastal launch window closes with a successful ascent.

    Spaceflight · resolves at window close

    12/12
    Pending
  5. 05

    WTI crude settles inside its Monday range every day this week.

    Energy · resolves Fri 1:30 PM CT

    10/12
    Pending
  6. 06

    The named polling average moves less than one point by Friday.

    Elections · resolves Fri 4:00 PM CT

    10/12
    Pending
  7. 07

    The weekend's #1 film holds the top of the domestic box office.

    Culture · resolves Sun estimates

    11/12
    Pending
  8. 08

    The 10-year Treasury yield ends the week below its Monday close.

    Rates · resolves Fri 2:00 PM CT

    10/12
    Pending
  9. 09

    The announced flagship device reaches general availability without a slip.

    Technology · resolves Fri 4:00 PM CT

    10/12
    Pending
  10. 10

    The favorite covers the spread in Sunday's late game.

    Sports · resolves Sun final

    10/12
    Pending
Consensus is the count of experts agreeing at the gate. Sample rows — the live board, its record and its receipts begin at launch.

How the board works

The engine Circle of experts

Independent experts

Multiple models, different architectures, different methods, worked in isolation. No expert sees another's answer, so agreement can only come from the evidence.

Live intelligence

Every expert queries current data mid-deliberation — entity corpora, prices, records, filings — through our sister platform, Verumdare.

The convergence gate

Predictions are published only when the circle lands on essentially the same outcome with high confidence. Everything else is discarded, unpublished.

Public judgment

Friday grades the week. Hits, misses and pushes go on the permanent record — because a forecaster who buries misses is an entertainer.

Inside the engine

Against the crowd

A market prices the crowd's opinion.
We publish the circle's verdict.

Prediction markets such as Kalshi and Polymarket are genuinely useful instruments — they aggregate what people with money at risk believe. But a market's answer moves with sentiment, liquidity and herding, and it never has to explain itself.

The board is a different animal: independent experts, evidence pulled live from the record, a hard gate, and a graded history. We think the two will end up in conversation — the circle's verdict against the crowd's price is itself a signal, and putting the engine's convictions to work on regulated venues is on our roadmap.

Predictions are research output, not investment advice — the board is a scoreboard, not a brokerage.

The Monday email

One email when each week's board publishes, and one when it is judged. No sequences, no filler.

Get the board

Be on the list when the live board starts publishing — Mondays at 9:00 AM Central, judged Fridays at 4:00.