Event probability
“Has Bitcoin bottomed?” gets one probability at the timestamp we form it. Not short, medium, and long versions of the same question.
Research
Data first. Then a clear opinion.
We publish event probabilities as of a timestamp, path forecasts for targets and trends, and the evidence and falsifiers behind both. Positions stay private. Reasoning does not.
Full research contract: How we form market opinions.
See the difference
One probability for events. Bands for paths.
BTC cycle low is inAs of latest book · residual 49%
4-year cycle still usefulLean yes · prior + position
BTC · medium · 90d pathbear / base / bull — not event odds
$46,926spot $65,175$85,000
Short / medium / long horizons never rewrite the event probability. They only describe possible price paths and trends.
Path bands
Short / medium / long are for where price might go — bear, base, and bull scenarios with weights. Growth and trends live here.
Falsifier first
Every claim writes what would prove it wrong before the outcome. We score later so opinions stay honest — not just narrative.
- 01Observe
Public market data
- 02Event P
One number as-of now
- 03Path
S / M / L targets
- 04Score
Brier · path hit-rate
- 05Learn
Priors · lessons · notes
- 06Act
Gated designated rails
The standard
Four questions, asked of everything.
01
What does the data say?
Public market feeds and portfolio multi-lens scores first — prices, dominance, macro proxies, cluster risk.
02
What is our event probability?
Binary claims get one probability as of the timestamp (e.g. “cycle low is in”). Not short/med/long odds.
03
What is the path?
Short / medium / long horizons are for price bands, growth, and trends — with scenarios, not fake precision.
04
What would prove us wrong?
Every claim carries a falsifier before the outcome so the system can learn, not just narrate.
Published
Reports
8 published
Research
Sapphire Security Digest — 2026-07-29
Weekly Sapphire security digest — 5 actively exploited CVEs from CISA KEV. As of 2026-07-29, four of five were past their CISA remediation dates. Authoritative CISA and NVD records are linked; informational only.
2026-07-29 · 3 min · read →
Research
Robinhood Agentic free-reign — how the desk is allowed to act
The public contract for free-reign easy on designated RH Agentic capital: clip-to-cap, per-venue slots, Super Heavy as planner only, kill switch always wins.
2026-07-27 · 2 min · read →
2026-07-29
Sapphire Security Digest — 2026-07-29
Weekly Sapphire security digest — 5 actively exploited CVEs from CISA KEV. As of 2026-07-29, four of five were past their CISA remediation dates. Authoritative CISA and NVD records are linked; informational only.
securitycvethreat-intelkev
3 min
2026-07-27
Robinhood Agentic free-reign — how the desk is allowed to act
The public contract for free-reign easy on designated RH Agentic capital: clip-to-cap, per-venue slots, Super Heavy as planner only, kill switch always wins.
tradingagenticrobinhoodfree-reign
2 min
2026-07-27
Learning loop — wins, losses, calibration
Scheduled scoring of event probabilities and path forecasts. Brier calibration, lessons, and prior updates for the self-improving desk.
calibrationlearningconjecturepost-mortemmethodology
1 min
2026-07-27
Market opinions — events and path forecasts
Point-in-time event probabilities, short/med/long path forecasts, data quality, drivers, and written falsifiers.
conjecturecryptomacrocommoditiespredictions
9 min
2026-07-27
Portfolio research — multi-lens book review
Predictions and analysis for held names and priority interests — fundamental, quant/TA, narrative/vibes, related bluechips, OSINT situations.
portfolioholdingsquantfundamentalnarrative
17 min
2026-07-27
How we form market opinions
Event probabilities as of now, path forecasts for targets and trends, public data, falsifiers, and a scheduled self-improving learning loop that scores wins and losses.
methodologyconjectureresearchlearningcalibration
3 min
2026-07-25
How research is published here
The standard every report is held to, what is deliberately withheld, and why an empty page is better than a confident wrong one.
methodologystandards
3 min
2026-07-25
What the Machine Room refused to invent
A measurement audit of the public system map, the telemetry claims it could support, and the gaps it now leaves visible.
engineeringtelemetryaudit
5 min
What is withheld, and why
Current holdings, sizing, and exact capital figures are excluded from this corpus. The public architecture feed is a separate contract and contains none of them; a real-time public book would invite front-running.
What that leaves is the part worth reading anyway: the framework, and closed calls scored against falsifiers committed in advance. A method you can apply yourself is more useful than a ticker you cannot check.