01
Prices it never saw
An idea is tuned on one stretch of prices and judged on the next, which it has never seen. A rule that only works on the prices it was fitted to is a description of the past, not an idea.
Signals · tested before they are shared
Signals, tested first.
A signal here starts as an idea, written down with its reason before it runs, tested on prices it has never seen, and published pass or fail. Most fail. You see those too.
- Written down first
- Tested after costs
- Published pass or fail
- Never advice
33tested0passed
- Cross market · 26 Oct 2026Metals dollar confirmTurned down: score below the pass mark
- Cross market · 10 Oct 2026Silver leads goldTurned down: score below the pass mark
- Cross market · 10 Oct 2026Equity stress havenTurned down: too few trades to judge
- Session · 8 Oct 2026New York opening-range breakoutTurned down: cannot be told apart from luck
Dear trader who has been sent a signal
Most signals tell you what. Never why.
“BUY GOLD NOW. SL 4,180. TP 4,260. 🔥”
You know the message. Sometimes it is right. Sometimes it is quietly deleted. You never find out which happens more often, because nobody is counting.
There is no reason attached. No record. No costs. And when it fails, there is no post about it, only the next one.
Here a signal has to earn the name. It starts as an idea.
An idea here is written down first, with its reason, before it runs. Nobody can rewrite it afterwards to fit what happened.
Then it is tested on prices it has never seen, with every cost of a real trade charged.
Then it is judged by the same fixed tests as every other idea, ours included. Most fail. The failures are published next to the passes, dated, in public.
Nothing is called profitable until the judge has passed it. Not by us. Not by a headline. Not by a message at 11pm.
That is slower than a message at 11pm. It is also the only version of this that tells you the truth.
The life of an idea
Written down. Tested. Judged. Published.
The same five steps for every idea, ours included.
- 01
Written down
The idea and its reason are recorded before the test runs, so nobody can quietly rewrite it to match what happened.
- 02
Tested
Tuned on one stretch of real prices and judged on a later stretch it has never seen. Eight times over. Spread, commission, slippage and financing charged on every trade.
- 03
Judged
Nine fixed tests, the same for every idea. The four that do most of the work are drawn below.
- 04
Published
Pass or fail, dated, in public. The failures sit next to the passes.
- 05
Watched
An idea that passed keeps being watched. If it stops working, the record says so.
Anatomy of an idea
What is written down before it runs.
Here is what a typical message leaves out, and what every idea here has on its page before a single test runs.
The market and the rule. What it looks at, what makes it enter, what makes it leave. In plain words, no hidden second rule.
The stop and the limits. Where it is wrong, how much it may lose on one trade and in one day. Set first, never after.
The reason. Why the idea should work at all. An idea with no reason is a guess with a chart on it.
The date. The page is stamped before the test. If the rule changes, it is a new idea with a new date, and the old one stays on the record.
- Rule, stop and limits in plain words
- A reason, written before the test
- Dated, so it cannot be rewritten later
- Market: gold, 15-minute candles.
- Enters: when the close breaks above the last 20 candles’ high.
- Leaves: at a stop set before the entry, or when the break fails.
- Limits: a fixed loss per trade and per day, written first.
- Reason: breaks that hold tend to run, and the test will say whether this one does.
- Stamped before the first test. Never edited afterwards.
What the judge checks
Four questions most signals never answer.
Nine tests in all.
These four do most of the work.
Fail one and the idea is turned down.
02
The whole test
Every idea is tested on six years of real prices with spread, commission, slippage and financing charged, judged by 9 locked tests on data it has never seen, and checked against 2,000 shuffled copies of itself to rule out luck. How we test
03
Every cost, then doubled
Spread, commission, slippage and overnight financing on every trade. Then the judge doubles them all. An idea that only works on a perfect day is not one.
04
Better than luck
Test enough ideas and one will look good by chance. The judge raises the bar for every idea tried, and checks the timing against random entries.
One idea, judged
Our best idea so far. Still turned down.
Most pages like this show you a made-up example. Here is a real one instead: the idea with the highest score on the record so far, exactly as the judge saw it.
New York opening-range breakout. It traded 448 times on prices it was never tuned on, across eight stretches of time, with the spread, the commission, the slippage and the financing charged on every trade.
It passed 8 of the nine tests. Watch them tick down the window. Enough trades to judge. A score above the pass mark. Still standing when every cost was doubled.
Then the luck check. Test enough ideas and one of them will look good by accident. The judge asks how good the best of that many tries would look if none of them had any edge at all. This idea did not clear that bar by enough.
So it was turned down. Published, dated, and left on the record next to the others.
A judge that passes its best idea anyway is a rubber stamp. This one said no, which is why a yes from it will mean something.
448 trades on prices it was not tuned on
- Enough trades to judge
- Score above the pass mark
- Survives doubled costs
- Enough half-years in the plus
- No single period carries it
- Timing better than chance
- Cannot be told apart from luck
- Neighbouring settings hold up
- No prices from the future
8 unseen stretches up or down
Turned down, judged 8 Oct 2026
Why ideas fail
Most ideas do not make it. Here is where they fall.
Anyone can show you the winners. The more useful page is the one that shows where the rest went wrong.
Of the 29 ideas that traded in the test, 0 passed every check.
The check that turned down most of them: cannot be told apart from luck, which 29 of the 29 failed. Most failed more than one.
A rule that looks clever on one stretch of prices is easy to find. A rule that survives unseen prices, every cost and a doubled bill is rare. That is the point of the test, not a flaw in it.
Every failure stays on the record, dated. A channel that shows only its wins is selling you the survivors.
29traded in the test4never switched on0passed
- Cannot be told apart from luck29
- Score below the pass mark28
- Fails when costs are doubled26
- One period carries the result25
- Timing no better than chance25
- Too few profitable half-years22
The whole record
33 ideas. Every score, in the open.
One line for each idea the judge has seen, best score first. The bar is its score after every cost, on prices it had never seen. The gold line is the pass mark. Open a line for the tests it failed.
1.01best score1.00pass mark0passed4never switched on
01New York opening-range breakoutSession · 8 Oct 2026score 1.01, turned down
- Kind
- Session
- Judged
- 8 Oct 2026
- Trades
- 448
- Verdict
- Turned down
Tests it failed
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
02Asian range breakout with a trend filterSession · 8 Oct 2026score 0.96, turned down
- Kind
- Session
- Judged
- 8 Oct 2026
- Trades
- 211
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
03Breakout after an inside dayVolatility · 8 Oct 2026score 0.86, turned down
- Kind
- Volatility
- Judged
- 8 Oct 2026
- Trades
- 163
- Verdict
- Turned down
Tests it failed
- Too few trades to judge
- Score below the pass mark
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
04Morning direction held to the New York closeSession · 8 Oct 2026score 0.77, turned down
- Kind
- Session
- Judged
- 8 Oct 2026
- Trades
- 775
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
05Morning direction, large moves onlySession · 8 Oct 2026score 0.73, turned down
- Kind
- Session
- Judged
- 8 Oct 2026
- Trades
- 423
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
06Daily momentumTrend · 7 Oct 2026score 0.73, turned down
- Kind
- Trend
- Judged
- 7 Oct 2026
- Trades
- 46
- Verdict
- Turned down
Tests it failed
- Too few trades to judge
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
07Break of yesterday's high, second versionTrend · 7 Oct 2026score 0.71, turned down
- Kind
- Trend
- Judged
- 7 Oct 2026
- Trades
- 479
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
08Volatility breakout from the day's openVolatility · 8 Oct 2026score 0.58, turned down
- Kind
- Volatility
- Judged
- 8 Oct 2026
- Trades
- 304
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
09Silver leads goldCross market · 10 Oct 2026score 0.45, turned down
- Kind
- Cross market
- Judged
- 10 Oct 2026
- Trades
- 1,387
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- One period carries the result
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
10Range expansion from the openVolatility · 7 Oct 2026score 0.45, turned down
- Kind
- Volatility
- Judged
- 7 Oct 2026
- Trades
- 357
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
11Metals dollar confirmCross market · 26 Oct 2026score 0.37, turned down
- Kind
- Cross market
- Judged
- 26 Oct 2026
- Trades
- 418
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
12Time-of-day patternCalendar · 7 Oct 2026score 0.36, turned down
- Kind
- Calendar
- Judged
- 7 Oct 2026
- Trades
- 905
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
13Pullback in a four-hour trendTrend pullback · 7 Oct 2026score 0.32, turned down
- Kind
- Trend pullback
- Judged
- 7 Oct 2026
- Trades
- 544
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
14Moving-average pullback with a VWAP filterTrend pullback · 7 Oct 2026score 0.29, turned down
- Kind
- Trend pullback
- Judged
- 7 Oct 2026
- Trades
- 477
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
15Asian range breakoutSession · 7 Oct 2026score 0.09, turned down
- Kind
- Session
- Judged
- 7 Oct 2026
- Trades
- 525
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
16Break of yesterday's high in an uptrendTrend · 8 Oct 2026score -0.01, turned down
- Kind
- Trend
- Judged
- 8 Oct 2026
- Trades
- 150
- Verdict
- Turned down
Tests it failed
- Too few trades to judge
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
17Hourly channel breakoutTrend · 7 Oct 2026score -0.03, turned down
- Kind
- Trend
- Judged
- 7 Oct 2026
- Trades
- 232
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
18Breakout after a quiet squeezeVolatility · 7 Oct 2026score -0.10, turned down
- Kind
- Volatility
- Judged
- 7 Oct 2026
- Trades
- 206
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
19New York metals-open breakoutSession · 7 Oct 2026score -0.13, turned down
- Kind
- Session
- Judged
- 7 Oct 2026
- Trades
- 763
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
Its line on the record →8 unseen stretches
20Momentum after 08:30 New York newsNews · 7 Oct 2026score -0.14, turned down
- Kind
- News
- Judged
- 7 Oct 2026
- Trades
- 254
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
21Equity stress havenCross market · 10 Oct 2026score -0.23, turned down
- Kind
- Cross market
- Judged
- 10 Oct 2026
- Trades
- 132
- Verdict
- Turned down
Tests it failed
- Too few trades to judge
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
22Gold against its dollar and real-yield valueRelative value · 8 Oct 2026score -0.31, turned down
- Kind
- Relative value
- Judged
- 8 Oct 2026
- Trades
- 60
- Verdict
- Turned down
Tests it failed
- Too few trades to judge
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
23Many weak signals in one modelEnsemble · 8 Oct 2026score -0.32, turned down
- Kind
- Ensemble
- Judged
- 8 Oct 2026
- Trades
- 253
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
24London direction carried into New YorkSession · 7 Oct 2026score -0.42, turned down
- Kind
- Session
- Judged
- 7 Oct 2026
- Trades
- 601
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
25Fading sharp 15-minute movesMean reversion · 7 Oct 2026score -0.42, turned down
- Kind
- Mean reversion
- Judged
- 7 Oct 2026
- Trades
- 448
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
26New York session, long only (control)Control · 8 Oct 2026score -0.60, turned down
- Kind
- Control
- Judged
- 8 Oct 2026
- Trades
- 256
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
27Narrowest day of seven, then the breakVolatility · 8 Oct 2026score -0.66, turned down
- Kind
- Volatility
- Judged
- 8 Oct 2026
- Trades
- 20
- Verdict
- Turned down
Tests it failed
- Too few trades to judge
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
28Fading the edges of a quiet rangeMean reversion · 7 Oct 2026score -0.87, turned down
- Kind
- Mean reversion
- Judged
- 7 Oct 2026
- Trades
- 244
- Verdict
- Turned down
Tests it failed
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
29Reversal after the afternoon gold fixFixing · 7 Oct 2026score -0.95, turned down
- Kind
- Fixing
- Judged
- 7 Oct 2026
- Trades
- 160
- Verdict
- Turned down
Tests it failed
- Too few trades to judge
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Its line on the record →8 unseen stretches
30Asian session mean reversionMean reversion · 7 Oct 2026never switched on, turned down
- Kind
- Mean reversion
- Judged
- 7 Oct 2026
- Trades
- 0
- Verdict
- Turned down
Its line on the record →Tests it failed
- Never switched on: no setting earned in training
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
31Asian session sweep and reversalStop hunt · 7 Oct 2026never switched on, turned down
- Kind
- Stop hunt
- Judged
- 7 Oct 2026
- Trades
- 0
- Verdict
- Turned down
Its line on the record →Tests it failed
- Never switched on: no setting earned in training
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
32Fading yesterday's high and lowStop hunt · 7 Oct 2026never switched on, turned down
- Kind
- Stop hunt
- Judged
- 7 Oct 2026
- Trades
- 0
- Verdict
- Turned down
Its line on the record →Tests it failed
- Never switched on: no setting earned in training
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
33Two-day channel breakoutTrend · 8 Oct 2026never switched on, turned down
- Kind
- Trend
- Judged
- 8 Oct 2026
- Trades
- 0
- Verdict
- Turned down
Its line on the record →Tests it failed
- Never switched on: no setting earned in training
- Score below the pass mark
- Fails when costs are doubled
- Too few profitable half-years
- One period carries the result
- Timing no better than chance
- Cannot be told apart from luck
- Too few of its settings make money
Scores after spread, commission, slippage and financing, on prices each idea was never tuned on. Clearing the pass mark is not enough: an idea has to pass all nine tests, the luck test included. Past prices only, never a promise.
Read a result
Thirty seconds. Five things to look at.
Every line on the record reads the same way. Here is how to read one.
- 01
The name and the date
What the idea was and when it was judged. A result with no date is a story.
- 02
The trades
How many times it traded. Few trades can only ever be a hint, so the record says too early, and means it.
- 03
The score after costs
One number, with spread, commission, slippage and financing already taken off every trade.
- 04
The failed check
If it was turned down, the line says which test did it. That is usually the most useful part.
- 05
The verdict
Passed, turned down or too early. Never a promise about next week.
The record so far
The numbers, unflattering ones included.
33
ideas tested by the judge so far
0
passed every check
8
stretches of unseen prices each idea is judged on
2×
every cost, charged twice, to see if the idea survives a worse day
Counted from the research journal. Updated 26 October 2026. A pass is not a promise of what comes next.
Where the ideas live
Four places. One standard.
An idea is not a message. It is a page you can go back to. Here is where to look.
The record. Every idea Baroq has tested, with its score, its costs and the test it failed if it failed. At the last update, 33 ideas had been tested and 0 had passed. We print that number even when it is unflattering. Especially then.
The method. How the judge works, written down before the first idea ran, so the bar could never be moved to fit an idea someone liked.
Your own agent. Say your idea in one sentence. The agent writes the rules, the Lab tries them on past candles, and then the agent runs them on paper, on live prices, with a stop on every trade and a reason for every decision. Baroq’s own ideas face the locked judge, which is harder still.
The desk. If you already run robots, strategies or a channel on MetaTrader 5, the desk reads your own account and gives each one its verdict after every cost: promising, too early, or no edge. Your ideas, judged by the same standard, on your own trades.
An idea that passes earns a place. An idea that fails earns a line on the record. Both are worth more than a message you cannot check.
The difference
A signal channel asks for trust. Ours shows its work.
| A typical signal channel | A signal here | |
|---|---|---|
| Says “buy now” with no reason | Yes | No |
| Shows the reason | No | Yes |
| Written down before it runs | No | Yes |
| Tested on prices it never saw, after costs | No | Yes |
| Published when it fails | No | Yes |
| Is advice | No | No |
Who this is for
If one of these is you, start with the record.
Not everyone wants to read why an idea failed. Here is who usually does.
You have paid for signals before. You know the feeling of scrolling up to check a call and finding it gone. Nothing here gets deleted. The ideas that failed stay on the page, with the date and the reason.
You have an idea of your own. A rule you trust, never tested properly. Write it in a sentence and the Lab runs it on past candles, then your agent runs it on paper. You read the result before anything is near your account.
You follow a channel and wonder. Put its trades through the desk. They get a verdict after every cost, the same way everything else here does.
You want the truth more than the thrill. Good. Most ideas fail this test. That is the process working, not a bug.
If none of these is you, that is fine too.
The free tools work out position size, risk per trade and drawdown on one page. No email needed.
Read this before you read the ideas
An idea is not advice.
We would be doing you a disservice if you read the record as a shopping list.
A passed idea is not a promise. It survived a hard, honest test on prices that already happened. Markets change. An idea that passed can stop working, and the record will say so when it does.
A failed idea is not worthless. Often it teaches more than a pass. It shows you what looks clever and is not, before you pay to learn it yourself.
Nothing here tells you what to buy. Not the record, not a verdict, not an agent’s log. It shows you what an idea did. What you do next is your decision, on your account, with your broker.
No signal here says “buy now.” No line on these pages tells you to act. Anything shared on Telegram is for information only, never advice.
What the ideas do give you is a way to see, in the open, which rules survive a fair test and which do not.
That is more than a signal channel ever offered.
The offer
Read the record free. Test your own idea free.
No call to book. No sales pitch. The record and the method are open to everyone, and your first agent is free.
- 01
Step one · free
Test your own idea
Say it in one sentence. Your agent writes the rules and runs them on paper.
- Tried on past candles in the Lab first
- A stop on every trade
- A reason for every decision
- Just your email. No card.
- 02
Always free · for everyone
The record and the method
Every idea Baroq has tested, and how the judge works.
- Pass or fail, dated
- The score, the costs, the check it failed
- The method, written before the first test
- Updated as ideas are judged
Information only, never advice. Software only: Baroq never holds your money and never trades for you.
The small print, up front
Cancel any month. First payment back in 14 days.
No result is promised on this page, so no result is guaranteed. What you get is the work in the open, and your money back if you want out.
Cancel any month. The first payment is refunded in full if you ask within 14 days, no reason needed.
Questions
What people ask about a tested signal.
Are these signals like a channel sends?
No. A signal here starts as an idea written down with its reason, tested on prices it never saw and published pass or fail. Nothing on these pages tells you what to buy, and anything shared on Telegram is for information only, never advice.
What is an idea here?
A set of trading rules written down before it is tested: the market, the entry, the stop, the limits and the reason. Its result is published whether it passes or fails.
Why do most ideas fail?
Because the test is hard on purpose. A judge that passes everything is a rubber stamp.
Is a passed idea a recommendation?
No. It means the idea survived an honest test on prices that already happened. Markets change. What you do with that is your decision.
Can I test my own idea?
Yes. Say it in one sentence. The Lab tries it on past candles and shows the part it never saw on its own, with a luck test beside it. Then your agent runs it on paper on live prices. Baroq’s own ideas face the locked judge; yours get the Lab; neither is called profitable before it proves itself. See how an agent works
Is any of this advice?
No. Everything on these pages is shared for information. Baroq is software, not an adviser, and it never knows what gold does tomorrow.
Where is the failed ideas list?
On the record, next to the passed ones, dated. Open the record
How many ideas have been tested?
33 at the last update, and 0 passed every check. The count comes from the research journal, not from this page.
What is the luck test?
Test enough ideas and one will look good by chance. The judge asks how good the best of that many tries would look if none had any edge, and raises the bar to match.
Why are costs doubled?
A test on a perfect day proves little. If an idea only works when spread, commission, slippage and financing are cheap, it is not one you can lean on.
Can an idea that passed stop working?
Yes. Passed ideas keep being watched, and the record says so when one stops working. A pass is a result on past prices, never a promise.
Does Baroq place trades for me?
No. Baroq is software. It never holds your money and never places, changes or closes a trade.
Can I cancel?
Yes, any month. The first payment is refunded in full if you ask within 14 days.
One more thing
You can keep trading messages you cannot check.
Or you can read what every idea really did, after every cost, in the open.
Written down. Tested. Judged. Published. In that order.
Keep reading
Testing an idea? Start here.
Guides for testing an idea
- How to backtest a trading strategy for free (and trust the result)Free ways to backtest a trading idea: the MT5 tester, a chart platform, a spreadsheet or code, and the checks that make a free test honest.
- How to test a trading bot before using real moneyA four-stage plan to test a trading bot before real money: read the rules, backtest with costs, run on demo, then start tiny. What to watch and when to stop.
- What a real gold EA backtest should includeA gold EA backtest means little without real costs, good tick data, years of different markets and unseen data. What to ask for before you trust one.
- Is AI trading legit? How to tell real software from hypeAI trading software can be legitimate, but many promises are not. Seven checks that separate honest tools from hype, and what no AI can promise you.
- How to spot a martingale gold EA before it costs youMartingale and grid gold EAs look smooth until one trend wipes the account. How to spot them on MT5 from the record, the settings and the drawdown.
- Trading journal for gold: what to log and how to review itWhat to record for every gold trade, the fields that matter (setup, risk, costs, mistakes), a weekly review routine, and how to see if a rule really earns.
Pricing
Three ways in. Pick yours.
See all plans · Try one agent free
Baroq never holds your money and never trades for you. Educational only, not financial advice; trading involves risk.
01 of 08 · Home
Like steroids for trading.Your idea tested by an AI agent, our signals judged in public, and your whole desk scored after every cost.→02 of 08 · AI agents
Your idea, tested on real prices.An AI agent writes the rules back to you, in words you can actually read.→03 of 08 · Desk
Know what earns. To the cent.Every robot, strategy and signal on your MT5 accounts, judged after spread, commission and swap.→You are here · 04 of 08 · Signals
Signals that prove themselves.Each idea written down with its reason, tested on prices it never saw, published pass or fail.→Next · 05 of 08 · Free Tools
Know the cost before you trade.Lot size, contract value and a trade plan you can print. Your numbers in, a clear answer out.→06 of 08 · Education
Check any gold EA before you pay.Plain guides on size, risk and drawdown, and the tricks that make a weak record look strong.→