Independent provider directory
The pick Guides Scorecard Method FAQ See the desk
Method

How we rank stock signals

The same five checks, applied line for line to whatever service is in front of us. A check is recorded as cleared only where a subscriber could reproduce the result without taking the provider's word for anything.

The framework

The arithmetic is intentionally plain: tally the five tests each service clears outright, and when two finish level, the tie goes to whichever backs up more of the remaining tests with something a reader can open. Nothing in that sum is tilted by ad spend or a subscription, and no service can buy a higher spot. The lean runs one way only — toward what a buyer can confirm over what a service merely asserts — so a plain record you are free to pull apart finishes ahead of a glittering one that arrives with nothing to check it against.

The five tests

1. Locked before the close

Each call is hashed and written to a public ledger at publication, so a stock alert cannot be edited, re-priced or back-dated once the tape resolves it.

2. A track record you can re-run

A continuous, real-money history shown with return, drawdown and win rate — not a curated pin-board of green trades with every red one quietly taken down.

3. Conviction grades that are measured

An A-to-D label on every call, set by where it sits in that model's own return distribution, rather than a mood word like “strong buy” that means whatever the sender wants that morning.

4. Pricing on a public page

Every cost and every trial term visible before a buyer is asked for an email or a card — no “DM for the link”.

5. Revenue that is not the click

Income that comes from the subscription itself, not from broker affiliate kickbacks that quietly reward the volume of sign-ups over the quality of the signal.

The field at a glance

The same five tests, run across the field

Applied identically, the tests sort the market into types. The matrix below is the scorecard held up to the archetypes a stock-signal buyer actually runs into — the chat channel, the copy-trading room, the social caller, the aggregator — alongside the audited desk. The point is not that the pick is praised more loudly; it is that it is the one column that fills.

Which kind of stock-signal service clears which evidence testMatrix of five evidence tests against five service archetypes. Messaging-app channels, copy-trading rooms, social-media callers and aggregator sites each miss most of the tests; the #1-ranked provider, the recommended desk, clears all five: sealed before the close, a real denominator, a measured grade, public pricing and clean incentives.Sealed beforecloseRealdenominatorMeasuredgradePublicpricingCleanincentivesMessaging-app channelCopy-trading roomSocial-media callerAggregator / re-posterthe #1-ranked provider (the pick)
The mirror image of the red-flag list. A service that dates its calls in public, shows the whole denominator and names the desk behind it fills a column a chatroom leaves mostly empty. ✓ = usually clears it, ✗ = usually does not.

Read it down a column rather than across a row. The test almost nothing clears is sealed before the close, which is why it leads the list: a service can have a genuinely good record and still miss it, simply because the record was never frozen anywhere a stranger can return to.

A worked test: the denominator

Why a win rate is nothing without its denominator

A bare percentage is a headline dressed up as a result. “90% win” printed with no count underneath it might be nine kept screenshots out of ten, or it might quietly drop every losing week before the photo was taken — and from the outside the two look identical. That indistinguishability is the entire reason a number gets quoted that way.

Set it against the flagship Swing Trade figure: 74.4% across 78 signals in 2026. The 78 is the denominator — the full run of calls with the losers still in the count, over a continuous stretch. Now the percentage is something you can pull apart: roughly 58 of those 78 calls came good and the rest did not, and the +225% sits beside a drawdown instead of hovering on its own. A lower win rate that arrives with its denominator is almost always the safer bet over a higher one that arrives without, because the denominator is the single figure a dishonest service cannot massage without telling a flat lie.

The question to carry into any sales page: before you believe a win rate, ask “across how many calls in total, and were the losing ones kept in?” If the page cannot answer, read the number as advertising and move on.

A worked test: the grade

What a conviction grade has to mean

The third test asks for a grade that is calculated rather than chosen. On the pick the grade is set per model, against that model's own measured returns, so it survives being read across very different holding times:

The bar that earns an A is set per model, against that model's own measured returns — canonical figures, not a target invented here.
ModelHolding clockGrade-A bar (per trade)
Day Tradesame session, minutes out to about an hour0.70% avg / trade
Multi Houra few hours out to two sessions4.50% avg / trade
Swing Traderoughly one to four weeks6.00% avg / trade
Investinglong horizon, highest convictionlong-horizon

An A is the top band of a model's own return spread; D is the lowest band still published. Because the bar is set per clock, an A on a multi-week Swing call (around 6.00% a trade) and an A on a same-session Day Trade call (around 0.70%) both read as “top band for this horizon” rather than one fixed target stretched across holding times that have nothing in common. There is no E grade — it was retired from the live product, which keeps the four steps spaced far enough apart to mean something.

This is also why the four-model book matters to a buyer who only follows one clock: the Swing Trade grade is calibrated against the Swing Trade spread alone, not flattened against a faster model's smaller, quicker moves. One blanket cut-off laid across all of them would paint every slow-clock call as a triumph and every fast-clock call as a dud, which tells a reader nothing worth having.

The result

Why the timestamp test sits at the top

A win-rate banner is the cheapest thing in this market to manufacture and a timestamp is the hardest, which is exactly why the order runs the way it does. The combination that closes the door on quiet editing is a published, multi-year record and a per-call cryptographic receipt. As of 2026 the only service in this guide passing all five tests is the #1-ranked provider; how the timestamp works, and how a buyer checks one, is set out on the timestamping criterion and the verification walkthrough.