What a Verified Trading Track Record Actually Requires

A track record is either something a stranger can audit or it is a marketing claim wearing a number. This is what separates the two, why most cited win rates fail the test, and how to check any record yourself in about ten minutes. Research and education only — not financial advice.

A trading track record is either verifiable or it is a claim. Those are the only two categories. A screenshot of a green day, a quoted win rate, a "we called it" recap posted after the candle closed — none of these are records in any sense that would survive scrutiny. A genuinely verified trading track record is something a stranger can reconstruct from public evidence without trusting you at all. That standard is unforgiving, which is exactly why so little of what gets sold as a track record actually meets it.

The four things a verifiable record requires

Verifiability is not a vibe or a badge you buy. It reduces to four concrete properties, and a record that is missing any one of them collapses back into a claim.

Timestamps that precede the move. The call must exist, in public, before the price action it claims to have caught. Platform-native timestamps — a Discord message, a dated post — not "as posted earlier" screenshots and not edited messages. If the evidence only appears after the move, it is a recap, and recaps prove nothing except that hindsight is legible.
Losses retained, not deleted. Every desk loses. The audit question is whether the losers are still on the board next to the winners. A history you can scroll that contains stopped-out trades and ideas that simply expired is a record; a highlight reel of green is an advertisement.
No human curation between the call and the scoreboard. If someone decides after the fact which trades "counted," the scoreboard measures the curator's judgment, not the strategy. A trustworthy record scores everything it posts by a fixed rule, including the embarrassing entries.
Auditable by a stranger. The final test: could someone who dislikes you, starting from the public record alone, arrive at the same numbers? If reproducing the scoreboard requires taking your word for anything, the record is not verified — it is asserted.

Notice that none of these four is a performance figure. Verifiability is a property of the record, not of the returns printed on it. Ugly numbers on an auditable record are worth more than beautiful numbers on an unauditable one, because only the first kind can be checked.

Why most cited win rates are unverifiable

A win rate is a fraction, and a fraction is meaningless without its denominator. When a service advertises "78% winners," the immediate questions are: 78% of what, scored how, over how many trades? Almost every cited win rate fails on at least one of these before you even reach the honesty question.

The denominator is usually invisible. You cannot see the trades that were quietly deleted, the "watching" posts that were retroactively claimed as wins, or the ideas that were never entered into the count at all. The definition of a "win" is usually undisclosed — a call scored at the mid-price can be a win on paper while every member who took it paid the bid-ask spread twice and lost. And the sample is usually far too small to mean anything: over five trades almost any strategy looks like genius or garbage, and services quote whichever window flatters them. A win rate also tells you nothing about the size of wins versus losses; a 70% win rate with tiny winners and huge losers is a losing strategy, which is why profit factor and expectancy matter more than the win percentage anyone leads with.

The red flags

Some claims are self-refuting. You do not need to audit them to know they will not hold — the number itself is the tell.

The 90% problem. A sustained win rate above roughly 90% on real, tradable signals is a red flag, not a selling point. Legitimate edges are thin and noisy; a headline win rate that high almost always means one of three things — the losers are being deleted, "wins" are scored at unfillable prices, or the sample is tiny and hand-picked. When the number sounds too good to be true, the record-keeping is usually where the trick lives.

Other tells cluster around the same instinct to look better than the process is: performance figures with no hypothetical, simulated, or paper label attached in the same sentence; screenshots offered in place of a scrollable public history; a scoreboard with no visible losing streak; vague calls like "NVDA looking strong" that can be claimed as wins regardless of outcome; and results that quietly reset whenever a bad month would show. For the fuller catalog of what failure looks like, see trading discord red flags.

How to verify a track record in 10 minutes

You do not need special access to audit most services. The evidence is public or it does not exist. Here is the ten-minute pass:

Minutes 1–3: find an original call in situ. Open the public channel or record and locate a specific trade in its original message. Confirm the platform timestamp sits before the move. If you can only find recaps, stop — there is nothing to verify.
Minutes 4–6: hunt for the losers. Scroll deliberately for stopped-out and expired trades. If the history is all green, the record is either curated or too young to evaluate. Either way you have your answer.
Minutes 7–8: check the labels. Take any headline number and find where it comes from. Is it labeled hypothetical, simulated, or paper in the same breath? Backtested and paper figures that are dressed up as real-money results are the most common quiet lie in this industry.
Minutes 9–10: test one call for fillability. Pick one winning card and ask whether a normal member could actually have gotten that price — reasonable spread, real volume — or whether the "win" lives only at an untouchable mid. A scoreboard and the member experience are sometimes two different products.

If a record passes all four, it is rare and worth a closer look. If it fails the first two, the rest does not matter.

What honesty looks like — including ours

Honest record-keeping is unglamorous by design, and we would rather show it than describe it. ClaudeQuantAlgo published a hypothetical backtest in which the raw scanner, traded blind with no filters, produced 161 simulated trades at a 46.6% simulated win rate, a 0.82 simulated profit factor, and roughly −2% expectancy per trade — simulated numbers, and unflattering ones, which is exactly why they are published rather than buried. That is what a real edge looks like before it is refined: a coin-flip win rate and a profit factor below 1.0, which in a backtest means the raw signal lost money.

The temptation, at that point, is to go grid-searching for a flattering variant. We ran a 21-variant grid; its best-looking cell showed +362 simulated units — and the desk's own audit rejected that cell, because a single ticker accounted for 61% of the profit and the result was not statistically significant. Publishing the number we threw away, and the reason we threw it away, is the point. A verified trading track record is not a collection of wins; it is a process you can watch make decisions, including the decision to reject its own best-looking result. Our cards are posted before the move with a trigger, TP1, TP2, a stop, and a time-stop, to a public timestamped paper/model record — no real money — where losses stay up and corrections happen in the open. The full workings, ugly numbers included, live at the public record. Run the ten-minute audit on us, and on anyone asking for your subscription.

Common questions

What makes a trading track record 'verified' rather than just claimed?

Four properties: calls timestamped in public before the move, losing trades retained on the board, no human curation deciding after the fact which trades count, and a record a stranger could reproduce without trusting you. Missing any one turns the record back into a marketing claim.

Why can't I trust a service's advertised win rate?

A win rate is a fraction with a usually-invisible denominator. You typically cannot see deleted trades, the definition of a 'win,' or the sample size, and win rate alone ignores the size of wins versus losses. Any figure not labeled hypothetical, simulated, or paper in the same sentence deserves extra skepticism.

Is a 95% win rate a good sign?

It is usually a red flag. Real edges are thin and noisy, so a sustained win rate that high most often means losers are being deleted, wins are scored at unfillable prices, or the sample is tiny and cherry-picked. Treat it as a reason to audit harder, not to subscribe faster.

Does a paper or model-desk record count as verified?

It can — a clearly labeled paper record with public, before-the-move timestamps is more auditable than an unverifiable real-money claim. The problem is never paper trading itself; it is paper results presented as though they were real money.

See the process with your own eyes. The desk posts trigger-based cards to a public, timestamped record — losses included — and published the backtest where its own raw scanner loses. The scoreboard is free to watch. Join the floor →

Free to join · paid floors optional · research and education only

Last updated 2026-07-11 · ClaudeQuantAlgo Research Desk · research and education only.

Disclosures. ClaudeQuantAlgo is a research and education community. Nothing on this page constitutes financial, investment, legal, or tax advice, or a recommendation to buy or sell any security or derivative. No profit promises are made, ever — trading stocks, options, and forex involves substantial risk of loss; options positions can lose 100% of their value. The public scoreboard reflects a model ("paper") desk — no real money. Past performance — real, paper, or simulated — never guarantees future results. Hypothetical and simulated results have inherent limitations and no representation is made that any account will or is likely to achieve similar profits or losses. ClaudeQuantAlgo is not a registered investment adviser or broker-dealer. You are solely responsible for your own trading decisions. Never risk money you cannot afford to lose.