Market Sentinel Gray-rhino watch · Plain English
Public market-risk intelligence · Updated daily

Built for savers, not seers.

Visible risks leave evidence before they become emergencies. Market Sentinel watches those clues, explains what they mean in plain English, and keeps the receipts: every forecast, source, mock trade, and later grade stays in the public record.

Research only · No real money · No recommendations

Evidence, not promises
The public desk

See the work, not just the pitch.

The system now leaves a public record across four distinct channels. Start with the one that matches your question: what happened, what it means, what is being tested, or what the mock rules actually did.

Research radar

The slow risks behind the daily tape.

Daily moves are useful only when they connect to a durable question. These briefs trace the visible, high-impact risks the desk is following across financing, energy, inflation, and policy.

Forecasting, with receipts

A prediction is only useful if it can lose.

The record does not grade good stories. It grades pre-written probabilities against pre-written rules. The score is shown with its nominal sample size, and every unresolved call stays open until its deadline is checked. Related calls may share evidence, so 38 grades are not 38 fully independent trials.

Explore every call  →
Before
Odds

A numeric confidence is filed before the outcome is known.

Boundary
Rule

The exact success condition and deadline are written in advance.

After
Grade

The outcome is resolved from named evidence, not remembered selectively.

Learning
Score

Calibration measures whether confidence matched reality over time.

How it works

A short loop, repeated in public.

The method matters, but it no longer needs to dominate the home page. Four linked steps turn incoming facts into an analysis that can later be checked.

01 · Observe

Verify the fact

Capture the figure, exact source, timestamp, and a second check.

02 · Interpret

Name the risk

Separate price confirmation from narrative and explain the saver-level consequence.

03 · Test

Log material claims

When a claim is genuinely probabilistic, give it a confidence, deadline, and objective rule.

04 · Revisit

Keep the lesson

Grade the result, preserve the miss, and improve the process rather than the judge.

The boundary

Public record, not public advice.

  • Portfolio metrics are computed by code; factual figures link to named sources. Interpretation is labeled.
  • No real money, broker, or live orders. The $100K book is a controlled practice environment.
  • No black-swan theatre. The desk tracks visible gray rhinos and admits what it cannot know.
  • No quiet rewrites. Calls, grades, corrections, and dated reports remain inspectable.
  • No recommendations. The work explains risk; it does not tell anyone what to buy or sell.
Start with the evidence

Read today’s work before reading about the system.

The daily archive is the clearest entry point. The operating method and the longer story — why gray rhinos matter more than crash calls — each have their own page.