The live engine record
What the engine did with real money, next to SPY and QQQ — published every session, win or lose.
The record
Three returns, one window, one start date — the engine’s real account and both benchmarks on the same basis. Δ is engine minus benchmark in percentage points; negative means behind.
Engine -0.3% since 07-08: trailed the Nasdaq-100 (QQQ 0.5%) by 0.8pp; trails the S&P 500 (2.5%) by 2.8pp. (Settled record through 2026-09-11; refreshed live from /api/perf on load.)
By calendar month
Each month's return chained from the settled series (the month-end cumulative against the prior month-end), engine beside SPY and QQQ. A month is a handful of sessions, not a verdict; the current month is partial until it settles.
| Month | Sessions | Engine | SPY | QQQ | Engine − SPY |
|---|---|---|---|---|---|
| July 2026 | 16 | −0.62% | +0.22% | −3.30% | −0.84% |
| August 2026 | 21 | +1.72% | +2.67% | +4.19% | −0.95% |
| September 2026 (to date) | 8 | −1.34% | −0.36% | −0.27% | −0.98% |
Settled through 2026-09-11. The first month starts at the 07-08 anchor; sessions held flat carry no engine point and are not graded.
Every exit, named
Every closed, engine-originated round trip since the 2026-07-08 anchor, winners and losers, newest first, with SPY over the same holding window. Percent only: no dollars, no share counts. The operator-placed trades the record backs out and the session it holds flat are excluded here too, and named in the API note.
The trade-level ledger publishes after the post-close settle pass.
/api/trades serves the same ledger as JSON, free, with the exclusion lists.
Curious what produces these lines? See the board the engine trades from →
This is the record. It is not the backtest.
Above: real orders, real account, funding-adjusted, from one start date. The sample is thin — the session count says how thin — and a thin sample is evidence of nothing in either direction. A receipt, not a result.
The other number is a simulation. The decade-replay figures elsewhere on this site are the 2017–26 research backtest of the ranking backbone: different period, different capital, no orders. They validate the ranking brain; they are not this account and not a forecast. How to read a backtest before you trust it →
When the engine trails, this page says so at the same size as anything favourable — the summary is written by the engine, not for this page.
What the number includes, and leaves out
Both paragraphs are fields on the API — they travel with the number, so the frame cannot be dropped on the way to a marketing page.
Proof of prior
Every market morning the published board — every scored name and its score (each day’s count is in the log below), no account data of any kind — is SHA-256 hashed into a public append-only log. A digest committed before the market resolved cannot be back-fitted to the outcome.
Precisely: it proves the calls were made in advance and never edited. It does not prove the returns above — those come from the brokerage account.
Reproduce a committed hash in your browser
Your browser fetches the log, rebuilds that day’s scoreboard and computes the digest locally — nothing uploaded, no key, no account.
The full log, on-page so machines read it without a second request. As JSON, with canonical-JSON rules and reference implementations →
Reading the commitment log from /api/board/proof.
For machines
Both records are public JSON — no key, no account, no email.
/api/perf— the record above, including every field this page renders./api/board/proof— the full commitment log, the canonical-JSON rules, and reference implementations in Python and JavaScript.
Wiring an agent to the board itself? Endpoints and the free MCP server are on the agents page.
Past performance does not predict future results. Percentages describe realized structure over a thin sample. Research publication, not investment advice.
Read the board, feed your agent, or run the engine
The same board, three ways: read it yourself (Scanner, $12/mo), feed it to your agent with the discipline this record trades by (Coil Live (MCP feed), $29/mo), or own the whole system (Engine, $79 once, ships disarmed).
Get the Engine — $79 See the free previewCoil is software you install and run yourself, with your own brokerage credentials and capital. It is long-only and not investment advice, not a managed account, and not a signal service. Leveraged ETFs, where the engine uses them, can lose value rapidly, including total loss. The figures on this page are the live engine record on a real brokerage account, rendered as served; every other performance figure on this site is a research backtest — point-in-time and survivorship-free, not live or client returns. Past performance does not predict future results.