Bellwether

NQ ● flat overnight ● live testing 24/7

A Kalman filter tracks large-speculator positioning in Nasdaq-100 futures from weekly CFTC COT reports. When positioning drifts hard enough in one direction, Bellwether trades that bias once per day with fixed risk — long and short alike. Backtested lookahead-free on real trade prints and walk-forward validated; the exact model and thresholds stay private — the live track record doesn't.

CFTC COT positioning
Kalman filtering
Walk-forward validated
Fixed-risk intraday execution

Signals

$100/mo

Daily bias + exact execution plan via API
Live event feed (entries, scale-outs, exits)
Dashboard view + one URL for Claude/LLM use
Cancel anytime

Full source code

$9,999 one-time

Complete strategy source (Rust) incl. the model
Backtest harness + parameters
Own it, run it, modify it — no lock-in
Delivered within 24h to your account email

read this first

Paper record. You can lose more than you paid us. These curves are paper-traded on our own data — no customer account has traded them. Part of the record before each sleeve shipped is a backfilled replay, tagged bf. Past results, simulated or live, predict nothing: several strategies on this site were killed by their own out-of-sample data, and we publish those too. Futures are leveraged and nothing here is investment advice.

All five points in full

These are paper records, not client money — Every live track record on this site is paper-traded on our own market data. No customer account has traded these strategies. Costs are charged on every fill, but a paper fill and a real one are not the same thing.
The record starts 1 June 2026, and part of it is backfilled — Rows dated before a sleeve actually shipped were reconstructed by replaying our archived tick data through the same code. They carry a bf tag wherever they appear. That is a replay, not a forward test — treat it as the weaker evidence it is.
Past results predict nothing — Not the backtest, not the live paper record, not the Sharpe ratio. A strategy that worked for two months can stop working the day you buy it, and several on this site have been killed by their own out-of-sample data. We publish those too.
You can lose everything you put in — Futures are leveraged. A single NQ contract moves $20 per point, and a bad day can exceed your margin. Losses are not capped at what you paid us — they are capped by your account balance, and then by your broker calling you.
This is not advice — We publish signals. We do not know your account, your risk or your circumstances, and nothing here is a recommendation to buy or sell. Buying a strategy does not make it suitable for you, and we are not in a position to judge whether it is.
If you switch on auto-execution, orders are placed without you — Renting signals places no orders at all. But if you separately link a broker account and activate a strategy on it, our engine polls that strategy every minute and moves your position to match — no confirmation, no second look, including while you are asleep. That is a standing instruction you give per strategy and per account, and you can withdraw it at any time from the dashboard. Until you do, it keeps trading.

Track record

Backtest

Backtest 2025-03-31 → 2026-02-26 · 0.9y — lookahead-free, fills on real trade prints only, 1 contract incl. commission + slippage

Returns

Net P&L+$21,447

Annualized P&L+$23,595

Profit factor2.00

Expectancy / trade+$100

Risk-adjusted

Sharpe3.69

Sortino6.74

Calmar11.69

Recovery factor10.63

Drawdown & risk

Max drawdown−$2,018

Worst trade−$504

Max consec. losses4

Trade stats

Trades214

Win rate60.3%

Avg trades / mo19.62

Payoff ratio1.32

Avg win+$333

Avg loss−$253

Live paper-trading

Live paper-trading on tickstream data since 2026-06-01

Live · since 2026-06-01

Net P&L+$5,825

Trades39

Win rate64.1%

Profit factor3.82

Expectancy / trade+$149

Max drawdown−$1,009

Date	Side	Entry	Exit	P&L	Reason
2026-07-16	LONG	29513.25	29513.25	−$5	BE
2026-07-16	LONG	29513.25	29525.75	+$246	TP_HALF
2026-07-16	LONG	29473.5	29498.5	+$496	TP
2026-07-16	LONG	29473.5	29486	+$246	TP_HALF
2026-07-15	LONG	29955.75	29955.75	−$5	BE
2026-07-15	LONG	29955.75	29968.25	+$246	TP_HALF
2026-07-14	LONG	29786	29786	−$5	BE
2026-07-14	LONG	29786	29798.5	+$246	TP_HALF
2026-07-14	LONG	29811.25	29786.25	−$505	SL
2026-07-13	LONG	29710.25	29735.25	+$496	TP
2026-07-13	LONG	29710.25	29722.75	+$246	TP_HALF
2026-07-03	LONG	29904.5	29904.5	−$5	BE
2026-07-03	LONG	29904.5	29917	+$246	TP_HALF
2026-07-03	LONG	29929.75	29904.75	−$505	SL
2026-07-03	LONG	29928.75	29941.25	+$246	TP_HALF
2026-07-03	LONG	29921.25	29946.25	+$496	TP
2026-07-03	LONG	29921.25	29933.75	+$246	TP_HALF
2026-07-02	LONG	30076.5	30101.5	+$496	TP
2026-07-02	LONG	30076.5	30089	+$246	TP_HALF
2026-07-01	LONG	30188	30188	−$5	BE

Live results are paper-traded on real tickstream market data, one NQ contract, with each sleeve's own commission and slippage charged on every fill — between $9 and $24.50 round turn depending on how it exits. Every record runs from 1 June 2026. Rows dated before the sleeve itself shipped are a backfill and carry a bf tag wherever they surface in the trade table; everything from the launch date on was traded forward.

Rebuilt 31 July 2026. The whole record was re-derived from our own archived tick path: every session the archive covers end to end was replayed through the agent's unmodified production code — the backfilled portion included, so it is no longer backtest output but the same code run over recorded prints. That corrected fills which had been booked at the trigger level instead of at the price that actually printed; on gap opens the difference reached 470 points. Sessions the archive cannot cover end to end keep their rows as recorded, because deleting a real trade over a hole in our recorder would flatter the record. On the sessions used as the control, the replay reproduces the live rows to the cent.

The backtest is lookahead-free: weekly CFTC data is anchored on its actual publication day and fills are simulated on real trade prints only — the live agent runs the identical model. Past performance, simulated or live, does not guarantee future results. Not financial advice.