How we rank options signals
The same five tests, applied the same way to every service — and a test only counts as cleared when a buyer could check it themselves rather than lean on the provider's say-so.
The scoring is deliberately plain. Tally the tests each service clears outright; where two finish level, the tie goes to whichever leaves more of its partial evidence open to inspection. No sponsor gets a thumb on the scale and no fee buys a kinder line — the exercise rewards what a reader can confirm over what a seller merely publishes. A quiet record with its workings on show therefore ranks above a glittering one that asks to be taken on faith. We judge each service only on what it discloses, and assert nothing about which instruments it trades.
The five tests
1. Sealed before the outcome
Each call is hashed and written to a public ledger at the moment of publication, so a defined-risk signal — levels and grade together — cannot be edited, re-priced or back-dated once the market resolves it.
2. A defined-risk structure
Every call is a clear level with a stated stop, so the worst case is bounded and known when the trade is opened rather than discovered when it goes wrong.
3. A measured conviction grade
An A-to-D label on every call, placed by where it sits in that model's own return distribution — a probability-style read, not a mood word like “strong” that means whatever the sender wants.
4. A re-runnable record
A continuous, real-money history a named outside party has reviewed, shown with return, drawdown and win rate, the losers left in — not a highlight reel.
5. Revenue that isn't the click
Income from the subscription itself, not from broker affiliate kickbacks that quietly reward volume of sign-ups over quality of signal.
The same five tests, run across the field
Applied identically, the tests do not just grade one service — they sort the market into types. The grid below is the scorecard set against the archetypes an options buyer actually runs into: the chat channel, the copy-trading room, the social caller, the aggregator, and the audited desk. The point is not that the pick is praised more loudly; it is that the recommended desk is the only entry carrying a tick in every box.
Read the grid down a column rather than across a row. The test almost nothing clears is sealed before the outcome, which is why it leads the list: a service can hold a genuinely strong record and still fail it, simply because nothing it shows you was committed anywhere a stranger could re-check before the market answered.
A win rate is worthless without its denominator
On its own a percentage is a slogan, not evidence. “90% win” printed with no count beside it might be nine of ten hand-picked tickets, or it might drop every losing month off the bottom of the page — and there is no way for a buyer to tell which, which is precisely why it is printed that way.
Set that against the flagship's read: 74.4% across 78 Swing Trade calls in 2026. The 78 is the denominator — the full tally, losers folded in, over a continuous run. Now the percentage is something a buyer can pull apart: about 58 of those 78 calls closed in profit and the rest did not, and the +225% return reads against a stated drawdown instead of floating loose. A lower win rate carrying its denominator beats a higher one without it almost every time, because the count is the one figure a dishonest service cannot pad without lying outright.
The question to carry into any sales page: “out of how many calls, and are the losers still in the total?” If the page cannot answer it, read the headline number as marketing and move on.
What the conviction grade has to mean
The third test asks for a grade that is computed, not chosen. On the pick the grade is set per model, against that model's own measured returns, so it survives being compared across cadences that have nothing in common:
| Published model | Cadence | Grade-A bar (per trade) |
|---|---|---|
| Swing Trade | roughly 7 to 28 sessions in a position (the flagship) | 6.00% avg / trade |
| Investing | long-horizon, highest-conviction calls | long-horizon |
| Multi Hour | about half a session to two sessions | 4.50% avg / trade |
| Day Trade | opened and closed inside one session | 0.70% avg / trade |
An A sits at the top band of a model's own measured returns; a D is the lowest band still published. Reading the bar per cadence is the whole point: an A on the flagship Swing model (near 6.00% a trade) and an A on a same-session Day Trade call (near 0.70%) both translate to “top band for this horizon” rather than a single target stretched across holding times that have nothing in common. There is no E grade; it left the live product so the four-step ladder keeps its meaning.
This is also why the four-model book matters even to a buyer who only ever follows one of them: the flagship Swing grade is calibrated against the Swing spread alone, not flattened against a faster model's smaller moves. One universal cut-off applied to all four would dress every slow call as a triumph and leave every fast call looking timid, which tells a buyer nothing about conviction.
The result, and where it leaves the field
Run the five tests honestly and one combination is rare enough to be decisive: an audited multi-year record and a per-call cryptographic receipt with the grade sealed inside it. As of 2026 the only service in this guide that clears all five is the #1-ranked provider. Each criterion that does the heaviest sorting has its own page — the mechanism behind the top test on sealed before the outcome, the bounded-loss requirement on a defined-risk structure, and the calibration of the ladder on grades that are measured.