Every signal below comes from the Trading Legends Council — ten legendary trading methodologies voting on the same chart. Every outcome is published: the wins, the losses, and the trades the council refused to take. Get each signal the moment it fires, free, on Telegram.
Council Signals
Every signal is issued by a ten-legend AI trading council and published here the moment it fires — with entry, stop, target and the full outcome record. Nothing is edited after the fact.
Free · instant push notification the moment a signal fires
EURUSD signal track record
EURUSD: 2 signals issued, 1 resolved — net result +2.9R, win rate 100.0%, profit factor ∞, average risk-reward 2.3.
GOLD signal track record
GOLD: 2 signals issued, 1 resolved — net result +2.0R, win rate 100.0%, profit factor ∞, average risk-reward 2.0.
BTCUSD signal track record
BTCUSD: 3 signals issued, 2 resolved — net result −2.0R, win rate 0.0%, profit factor 0.00, average risk-reward 2.9.
Legends
Latest signals
LIVEpublished automatically · also posted to our TelegramSignal history
full record — wins and losses alike · 9 signals| Issued (UTC) | Pair | Direction | Entry | Stop | Target | R:R | Consensus | Outcome |
|---|---|---|---|---|---|---|---|---|
| Jul 13 · 19:00 | GOLD | SHORT | 3,996.80 | 4,040.92 | 3,908.56 | 2.00 | 100% | OPEN |
| Jul 13 · 19:00 | EURUSD | SHORT | 1.13859 | 1.14012 | 1.13614 | 1.60 | 100% | OPEN |
| Jul 13 · 11:00 | EURUSD | SHORT | 1.14298 | 1.14456 | 1.1384 | 2.90 | 100% | WIN +2.90R |
| Jul 13 · 07:00 | EURUSD | NO TRADE | — | — | — | — | 0% | PASS |
| Jul 13 · 04:00 | BTCUSD | SHORT | 62,831.99 | 63,580.52 | 61,333.46 | 2.00 | 100% | OPEN |
| Jul 13 · 03:00 | GOLD | SHORT | 4,075.32 | 4,118.39 | 3,989.25 | 2.00 | 100% | WIN +2.00R |
| Jul 13 · 03:00 | EURUSD | NO TRADE | — | — | — | — | 59% | PASS |
| Jul 12 · 20:00 | BTCUSD | LONG | 63,743.32 | 63,003.42 | 65,150 | 1.90 | 100% | LOSS −1.00R |
| Jul 12 · 08:00 | BTCUSD | LONG | 63,945.11 | 63,828.14 | 64,500 | 4.74 | 76% | LOSS −1.00R |
$ and % figures assume fixed fractional risk on your settings (non-compounding) and are estimates, not account statements. Trading involves risk. Nothing here is financial advice.
Reading the table above
How to read these signals
- Pair & direction — LONG means buy, SHORT means sell.
- Entry / stop / target — entry is where the trade starts, stop is where it closes for a loss, target is where it closes for a win.
- Consensus — how strongly the ten-legend council agreed on the call. Hover the consensus bar (or the % in the history table) to see the full ballot: which legends voted for the direction, which voted against, and which chose to abstain — plus each legend’s own conviction (0–1) in its read.
- NO_TRADE — the council met and chose to stay out. Those are published too, because knowing when not to trade is half the record.
The numbers speak your language: set your account balance and risk per trade in the controls above and everything on this page converts to dollars or percent. On the default settings — a $10,000 account risking 3% per trade — a signal with a 1:2 target reads as “risk $300 to make $600.” No jargon required.
Six-month backtest · EURUSD 4h
We backtested the council — including the trades it refused
Before a single signal went live, we replayed six months of EURUSD 4-hour convenes bar by bar — the same engine, the same ten-legend vote, no hindsight. And here’s the part that should raise your eyebrows: this entire result was produced on free AI models — the budget tier. Top-tier models from Anthropic (Claude) or OpenAI (ChatGPT) can reason more carefully than a free model, and are expected to sharpen these results further; that upgrade is what the premium channel runs on. Here is everything the free tier produced — the wins, the losses, and the two out of three signals the council chose to skip.
What those six months actually looked like
The running total of every gate-approved trade, in your units. It was not a straight line — nine straight losses in February, one monster winner in March, a long grind through spring, then a strong June. This is what a 36.5% win rate with winners twice the size of losers feels like while it happens — and the dips are published with everything else.
On default settings the curve slides to −$2,691 by late February, recovers sharply in early March, peaks at +$4,083 in late June and ends the window at +$2,706. (Enable JavaScript to see the interactive chart.)
Hover or tap the line for individual trades. 63 of the 66 trades resolved inside the window; the 3 still open when it closed are not plotted.
The edge is knowing when not to trade
The council produced 208 signals in the window. Taking every single one of them would have wiped the account. The trade gate skipped 142 and took 66 — and that filter alone is the difference between a blown account and a +27% run. It is also why the live feed above publishes NO_TRADE calls: the passes are part of the record.
Same signals, same window, same risk per trade. The only difference is the discipline to pass.
Which voices earned their seat
Net result by the legends who backed each trade (voter combinations with at least 5 resolved trades in the backtest). Green earned, red lost — and yes, we show the red.
Small samples — read these as anecdotes, not rankings. The point is transparency: when a combination underperforms, it is published here, not hidden. (These are trades where only the named legends voted for the final direction.)
- Instrument & window: EURUSD, 4-hour timeframe, a six-month window replayed bar by bar in two consecutive runs.
- Same engine as the live feed: each signal is a full council convene — ten legendary methodologies vote, the verdict sets entry, stop and target.
- Free AI models only: every convene in this backtest ran on free-tier AI models — the same ones powering the free channel. The premium channel will run the identical council on top-tier paid models from Anthropic (Claude) and OpenAI (ChatGPT), which we expect to sharpen results further.
- Sequential, one position at a time: the “as traded” figures respect the open-trade gate, exactly like the live channel. No overlapping positions, no cherry-picking.
- Money math: fixed-fractional risk per trade, non-compounding. Dollar and percent figures use the settings shown (default $10,000 balance, 3% risk); 1R = the distance from entry to stop.
- Fills: hypothetical fills at the signalled levels; spreads, slippage and swaps are not modelled.
Backtested performance is hypothetical, shown for transparency and education. It does not include all real-world costs, and past or simulated results do not guarantee future returns. 3% risk per trade is an aggressive setting — scale it to your own tolerance. Trading forex, gold and bitcoin involves substantial risk of loss. Nothing on this page is financial advice.
Meet the Council: ten legends, one vote
Most "AI signals" are a single model guessing. A council signal only exists when ten independent methodologies — each modelled on a legendary trader's published playbook — analyse the same chart and vote:
Disagreement is the safety mechanism. When the votes split, the signal is a pass — and you can filter the live table above by any legend to see exactly how each one has voted.
The engine behind this council — the actual decision logic each legend runs, not just the vote count — is open source: read the code on GitHub.
How to check whether any signal provider is real — including us
Around nine in ten signal channels show you photoshopped wins and quietly delete their losses. You should audit this page the same way you'd audit any provider. Here's the checklist:
- Are losses published? Scroll the history above — losing trades are listed in red with their exact result in your own dollar terms. Nothing is deleted.
- Are skipped trades published? NO_TRADE calls stay in the record. A provider that only ever fires signals has no discipline to show.
- Is the win rate believable? Ours floats around the mid-30s to 40s percent, with winners roughly twice the size of losers. Anyone promising 85–95% is selling you a story — that math almost never survives a stop-loss.
- Are the timestamps real? Every signal is published the moment it fires, with its issue time. Alerts hit Telegram at the same instant, so you can verify against the chart yourself.
- Is the full history reachable? Every page of past signals is linked below the table — not a screenshot, the actual record.
- Is the engine itself inspectable? The Trading Legends Council that generates every signal on this page is open source on GitHub — read the actual decision logic yourself, not just our claims about it.
If a provider fails this checklist, walk away. If we ever fail it, walk away from us too.
Watch the record build in real time. Join the free channel and get every signal the second it fires — then judge us on outcomes, not promises.
Free signals vs. premium: what's the difference?
Everything on this page is free and stays free: the signals, the full track record, the Telegram alerts. The free channel is the product — not a teaser. And remember: everything you see here runs on free-tier AI models — the budget option.
A premium channel is coming for traders who want more: the same council running on top-tier paid models from Anthropic (Claude) and OpenAI (ChatGPT) — deeper reasoning per legend, expected to sharpen results further — plus earlier alerts, more pairs, and per-trade management updates (move stop, take partial, close). Same honest record — this page is exactly how the premium signals are generated, just on stronger brains. Join the free channel now and you'll be first to know when it opens, at the founding-member price.
Frequently asked questions
Are free forex signals worth following?
Some are — if the provider publishes a complete record. The industry's real problem isn't free vs. paid, it's hidden losses. Judge any channel (free or paid) by whether you can see every outcome, then test it on a demo account for a few weeks before risking money. This page exists so you can do exactly that.
How accurate are these signals?
In the six-month backtest the council won 36.5% of trades — and still returned +$2,706 (+27%) on the default $10,000 account, because average winners ($690) were twice the size of average losses ($330). Accuracy alone is a misleading number: a 90% win rate can lose money if the losses are big enough. The live record above updates with every resolved trade.
How do I get the signals on Telegram?
Tap "Get signal alerts on Telegram" anywhere on this page and join the channel — it's free and alerts push the moment a signal fires. No email, no registration.
What does NO_TRADE mean?
The council convened, analysed the chart, and voted to stay out. We publish those calls because skipping bad setups is where the edge comes from — in our backtest, taking every signal without the filter would have wiped the account, while the filtered record gained +27%.
What pairs and markets are covered?
EUR/USD, gold (XAU/USD) and bitcoin (BTC/USD), on 4-hour charts. New signals are generated around the clock as the council convenes every four hours per market.
Do I need a big account to use these signals?
No. Use the balance and risk controls above and every number on this page rescales to your own account — $500 or $50,000. The default view assumes $10,000 at 3% risk per trade; most followers risk 1–3% of their account per trade.
Every signal, free, the moment it fires.
No email. No card. Just the record, building in public.
Join the free Telegram channelSignals, statistics and backtests on this page are published for education and transparency; they are not investment advice or a solicitation to trade. Dollar and percent figures assume fixed-fractional risk on the settings you choose (default $10,000 balance, 3% risk per trade). Past and simulated performance do not guarantee future results. Never trade money you cannot afford to lose.