Not financial advice. PairsCopilot sells data: the output of our statistical model, published identically to every subscriber. We are not investment advisers or brokers — nothing here is a recommendation to buy or sell any security, and no figure is a prediction or guarantee of profit. Do your own research: every trade is your own decision, made with your own risk management, at your own risk. Trading involves substantial risk of loss, including loss of principal.
PairsCopilot

A disciplined pairs-trading model. Its output, in your inbox, every evening.

Every market day one statistical model scans a validated universe of equity pairs for stretched relationships and publishes what it finds — identically, to every subscriber, in dollars and days instead of sigmas. We sell the model's data. The research, the decision, and the risk management are yours.

Full report — $39/month
or $299/year (save 36%)

The daily report is data from our model — not financial advice, not a recommendation. Unsubscribe anytime; subscribing creates no advisory relationship. Cancel the paid plan anytime from the receipt email.

Or see today's free opportunity first →

Every signal we publish is tracked to its resolution in an append-only ledger — timestamped at publication, losses shown identically to wins, impossible to edit after the fact.

One model, one process

Pairs are admitted only after passing statistical qualification, and get benched the moment they stop passing. Entries, exits, and safety stops come from pre-set rules — the same rules every night, never a hunch.

Same data for everyone

An impersonal publication: every subscriber sees identical analytics at the same time. Nothing is tailored to you or your account. It's a scanner and a discipline, not an advisor.

Losses shown, always

Every published signal is tracked to its resolution and the results build in the open — stopped-out losers with the same weight as winners. The safety-stop cost is printed next to every modeled gain.

How strict the pipeline is

We publish process, not promises. Every pair fights its way onto the board and gets benched the moment it stops qualifying.

30
Pairs admitted
Each passed a multi-year walk-forward qualification bar before its first signal could publish.
0 admitted
Placebo control
The same admission bar is run on shuffled placebo pairs; a batch only counts if the placebo set fails. Most recent control: 0 of 29 placebos passed.
5 of 29
Benched last night
Pairs failing the nightly stationarity or half-life gate publish nothing that night — a validity gate, re-tested every scan.
1
Suspended by health rule
Repeated stop-level stress auto-suspends a pair; reactivation is manual, never automatic. 29 candidates wait in validation.

Counts are live from the production database. Subscribers see how every published signal resolves — losers with the same weight as winners — inside the report.

Two ways to get the daily report

Full report — $39/month
or $299/year (save 36%)

The daily report is data from our model — not financial advice, not a recommendation. Unsubscribe anytime; subscribing creates no advisory relationship. Cancel the paid plan anytime from the receipt email.

Terms of ServicePrivacyDisclaimer & RiskArborium App LLC · 16826 Harbor Falls Dr, Cypress, TX 77433