Research control room

Can a bounded strategy improve asset accumulation after costs?

The research asks whether explicit trading rules—and, separately, limited AI assistance—can improve crypto holdings after costs. Results are compared with passive holding and cash. Forecasts, simulated paper results, order previews, and live-account records are shown separately so that one kind of evidence is not mistaken for another.
Public research accessMethodology, model evidence, and evaluation rules are open to review. Coinbase connection, personal Strategy Setup, private Run Monitor, and deployment-scoped live results require sign-in and are offered under the $9.99/month member plan.
View pricing
Current experiment

What does the available evidence show?

The summary gives priority to recorded observations. Projected values, order previews, planned capital, and missing responses remain labeled separately from live-account results.

Model-to-deployment evidence path

How is a model set, tested, selected, and evaluated?

Evidence becomes stronger only as a fixed hypothesis survives later stages. A good historical or paper result can justify a new forward test, but it does not become live authority and is never copied into the live return.

Estimated performance against actuals

Original projection, recorded outcome, and passive benchmark

Control room · 1Stage and authorizeDrafts, capital scope, model mix, routes, stops, publication, and authority Control room · 2Observe the runRead-only heartbeat, decisions, previews, commits, blockers, and health Control room · 3Compare the roundLocked entrants, current leader, benchmarks, calibration, costs, and measured AI treatment Control room · 4Inspect live accountingDay 0 attribution, orders, fills, fees, units, flows, and matched benchmarks Control room · 5Review and reviseCompleted-round lessons, failure analysis, and versioned next-round candidates
Research lab

Which model hypotheses still deserve attention?

These controls are local exploration only. They do not alter the official paper experiment, Coinbase account, selected live strategy, or executor.

Focus models

Visual Evidence Explorer

Linked model evidence

Position Growth Thesis

The governing question is whether a strategy improves per-asset ownership versus passive holding after exchange costs and marginal AI review cost—not whether it trades frequently.

Digital Finance Thesis Map
Compare Actual Performance Inspect Historical Evidence Browse Model Catalog
Bounded deployment

What must be configured before money can move?

Account connection, strategy definition, order previews, live commit authority, and unattended executor authority are separate stages. Advancing one stage does not silently authorize the next.

Verify Coinbase Configure Strategy Open Run Monitor Open Live Results
Evidence and governance

Can every claim be traced to a source and formula?

Use the audit material to trace each claim to its source, study definition, responsible page, and scheduled data-producing job.

Research Questions and Evidence

Inspect Snapshots Compare Experiment Rounds Read Methodology