Live execution pilot · broker demo

T4T Macro IFVG — Execution Log

Every signal the strategy sent, and exactly what the broker did with it.

What this measures is execution fidelity — whether the trade your broker actually took matches the trade your strategy intended. Where the fill landed against the price you sent, whether the stop ended up the distance you asked for, whether a third really came off at each target.

It is not measuring whether the strategy makes money, and at 0 closed trades it could not. Every figure below is a measurement of the plumbing; the assessment section says plainly where that stops and what a real sample would need.

Awaiting first signal
executor connected, no strategy signal yet
21 September 2026, 08:35 Cayman
← all accounts
Right now
Position
flat
1 of 1 executors connected
Market
OPEN
last quote 1s ago
Executor
LIVE
owner-ea live, beat 3m ago (every 5m)
Last signal
none yet
none yet
Today
0 of 4
4 more allowed · the trading day is New York, resets 23:00 Cayman
The record
Executed
0
0 closed, 0 refused
Trades closed
0
none yet
Planned → filled
no strategy fills to measure yet
Net result
on 0 strategy trades. Account has moved -994.41 in total; every closure we recorded, tests included, accounts for +0.00 — ⚠ -994.41 unexplained
What changed

Since this started

Everything else on this page is current state. This is the sequence — ⭐ built from the trade record and the configuration history, so it cannot drift from what actually happened.

WhenWhat
21 Sep 05:11Nasdaq 2-minute started — VANTAGE:NAS100 on NAS100, 4 windows
21 Sep 05:11Nasdaq 3-minute retired
20 Sep 16:48Gold 2-minute started — OANDA:XAUUSD on XAUUSD, 4 windows
15 Sep 19:00Nasdaq 3-minute started — VANTAGE:NAS100 on NAS100, 7 windows
15 Sep 19:00Nasdaq 3-minute (retired) retired
14 Sep 19:00Nasdaq 3-minute (retired) started — CAPITALCOM:US100 on USTEC, 4 windows
Results

Every trade the strategy produced

Taken from the broker's own deal history — not from TradingView's messages, which carry one alert for both legs of an exit and so cannot tell a stop-out from a take-profit. Planned is the price your strategy sent; filled is what the broker gave.

TimeInstrumentWindowDir PlannedFilledSlip ExitRisk askedRisk taken HeldWhyR P&LRunning
No strategy signals executed yet. The pilot is armed and waiting for its first macro window.

Nothing here yet. 0 hand-fired test trades are excluded on purpose: they proved the machinery, and counting them as your strategy's results would flatter or damage the record without meaning either.

Every leg, trade by trade

⭐ The table above is the outcome; this is the working. For a strategy that scales out in thirds, the outcome alone hides most of what happened — when a third came off, when the stop moved and to where, what the remainder closed at. Straight from the executor's own records; nothing here is derived.

no trades yet
Price action

What the market did around each trade

Candles from your broker's own feed — the same one that filled the order, so a candle and a fill can never disagree. ⭐ Each chart is drawn hours after the trade closed and deliberately runs on past the exit, because the useful question is not where we got in. It is what price did next.

No settled trades to chart yet. A trade is drawn a couple of hours after it closes, so the picture includes what happened next.
By window

Which macro window is producing this

Your strategy already thinks in macro windows and your own Pine has a per-macro table, so this is the same cut taken from real fills. ⚠ At 0 closed trades none of these figures can carry weight — it is here because it starts accumulating now, and because a window that never fires is worth seeing long before its numbers mean anything.

WindowSignalsLast Closed WonTotal RMean R P&LState
03:50no signal yet
08:50no signal yet
12:50no signal yet
23:50no signal yet

4 of 4 windows have never produced a signal: 03:50, 08:50, 12:50, 23:50. That is not yet a fault — a window is an opportunity, not a schedule — but a window still silent after a few weeks is worth questioning. ⛔ No conclusion should be drawn from the columns above until each window has a sample of its own; one trade gives a window a 100% or 0% win rate, which is arithmetic, not information.

The ledger

Every message, in order

What the relay did with each message before the executor ever saw it. Times are the relay's, which is the only clock in this chain that is both trusted and shared. ⚠ A rejected message is a missed trade, and a missed trade looks exactly like a quiet market — so each one is listed rather than counted.

TimeFromInstanceMessageOutcomeWhy
nothing received yet
Configuration

The settings behind every number here

Times on this page are Cayman time. ⚠ New York is 1 hour ahead of Cayman at the moment, because New York observes daylight saving and Cayman does not. So a window labelled 09:20 NY is 08:20 your time today — and the same clock in January, when the two align.

⚠⚠ A TradingView alert freezes your inputs, the chart symbol and the timeframe at the moment you create it, and keeps firing that frozen copy until it is deleted and remade. So a trade does not belong to “the settings” — it belongs to whatever was frozen into the alert that produced it. Each block below is one such configuration, with the trades that came from it.

⚠ TradingView offers no way to read an alert's settings back, so these are declared, not read. Where the messages themselves prove something, it is marked evidenced; everything else is our word for it.

Nasdaq 2-minute nas2m
CURRENT 2026-09-21 → now

⭐ The author's NQ specification, adopted 2026-09-21 at his request after he reviewed the dashboard. 2-minute, four macros, SL Buffer 2.0 (his default is 1.95), breakeven After 2:1 — ⚠ NOT 1:1, which is gold only. ⏳ Needs a NEW TradingView alert: the old one froze the 3m inputs and cannot be edited into this.

0 trades came from this configuration.

SettingYour defaultRunning asHow we know
Chart and instrument
ChartVANTAGE:NAS100declared
Timeframe2mdeclared
Broker symbolNAS100declared
AccountVantage demo, 1:100, USD 100kdeclared
Macro windows
03:50 NYoffonno signal yet
08:50 NYoffonno signal yet
12:50 NYoffonno signal yet
23:50 NYoffonno signal yet
Trade geometry, as the messages describe it
Reward multiples, scale-out1:1 · 2:1 · 3:1, thirdsdeclaredawaiting a real signal ⚠ hand-fired tests are deliberately not counted — their geometry is ours, not the strategy's
Strategy inputs changed from your defaults
Enable Webhook Alerts?falsetruedeclared
Ticker SymbolNQM2026NAS100declared
Move SL to BreakevenAfter 1:1After 2:1declared
SL Buffer (×ATR14)1.952.0declared
Max Trades Per Day74declared
Max Trades Per Macro Window11declared
Execution limits (ours, not yours)
Risk per trade0.5 % of balancedeclared
Position cap5.0 lotsdeclared
Daily loss limit3.0 %declared
Concurrent positions1 declared
Gold 2-minute gold2m
CURRENT 2026-09-20 → now

⭐ ARMED 2026-09-20 18:55 UTC, owner's decision. demo_only=no, KillSwitch=false, risk 0.05%, MaxOnAccount=1, breakeven After 1:1, four macros 08:50 · 10:50 · 19:50 · 21:50 NY, cap 4. ⛔ THE CHART-TO-BROKER PRICE OFFSET IS STILL UNMEASURED. Broker's Friday close was 4377.66 bid / 4378.26 ask (spread $0.60); compare against OANDA:XAUUSD's Friday close. Under ~$0.30 is fine. ⚠ No Risk Shield watcher exists yet - MaxDailyLossPc=2.0 is a local backstop that cannot see the 6%-of-peak-equity floor.

0 trades came from this configuration.

SettingYour defaultRunning asHow we know
Chart and instrument
ChartOANDA:XAUUSDdeclared
Timeframe2mdeclared
Broker symbolXAUUSDdeclared
AccountUpcomers Thunderbolt 1M, real, 1:100declared
Macro windows
08:50 NYoffonno signal yet
10:50 NYoffonno signal yet
19:50 NYoffonno signal yet
21:50 NYoffonno signal yet
Trade geometry, as the messages describe it
Reward multiples, scale-out1:1 · 2:1 · 3:1, thirdsdeclaredawaiting a real signal ⚠ hand-fired tests are deliberately not counted — their geometry is ours, not the strategy's
Strategy inputs changed from your defaults
Enable Webhook Alerts?falsetruedeclared
Ticker SymbolNQM2026XAUUSDdeclared
Move SL to BreakevenAfter 1:1After 1:1declared
SL Buffer (×ATR14)1.952.0declared
Max Trades Per Day74declared
Max Trades Per Macro Window11declared
Execution limits (ours, not yours)
Risk per trade0.05 % of balancedeclared
Position cap50.0 lotsdeclared
Daily loss limit3.0 %declared
Concurrent positions1 declared
Nasdaq 3-minute nas3m
RETIRED 2026-09-16 → 2026-09-21

⛔ RETIRED 2026-09-21. Ran the 3m seven-window set the author gave on 2026-09-17, with his Pine default SL Buffer of 1.95. He reviewed the dashboard on 2026-09-21 and said Nasdaq should be his NQ spec instead: 2-minute, four windows, SL Buffer 2.0, breakeven staying at After 2:1. ⭐ Its seven closed trades stay on the record - they measured THIS configuration, and pooling them with the next one would compare two different strategies.

0 trades came from this configuration.

SettingYour defaultRunning asHow we know
Chart and instrument
ChartVANTAGE:NAS100declared
Timeframe3mdeclared
Broker symbolNAS100declared
AccountVantage demo, 1:100, USD 100kdeclared
Macro windows
00:50 NYoffonno signal yet
03:50 NYoffonno signal yet
09:20 NYoffonno signal yet
18:50 NYoffonno signal yet
19:50 NYoffonno signal yet
22:50 NYoffonno signal yet
23:50 NYoffonno signal yet
Trade geometry, as the messages describe it
Reward multiples, scale-out1:1 · 2:1 · 3:1, thirdsdeclaredawaiting a real signal ⚠ hand-fired tests are deliberately not counted — their geometry is ours, not the strategy's
Strategy inputs changed from your defaults
Enable Webhook Alerts?falsetruedeclared
Ticker SymbolNQM2026NAS100declared
Move SL to BreakevenAfter 1:1After 2:1declared
Execution limits (ours, not yours)
Risk per trade0.5 % of balancedeclared
Position cap5.0 lotsdeclared
Daily loss limit3.0 %declared
Concurrent positions1 declared
Nasdaq 3-minute (retired) ustec3m
RETIRED 2026-09-15 → 2026-09-16

Retired: that broker refused every index order with retcode 10017. No trade ever executed on this epoch.

0 trades came from this configuration.

SettingYour defaultRunning asHow we know
Chart and instrument
ChartCAPITALCOM:US100declared
Timeframe3mdeclared
Broker symbolUSTECdeclared
AccountMetaQuotes demo, USD 100kdeclared
Macro windows
09:20 NYoffonno signal yet
14:50 NYoffonno signal yet
18:50 NYoffonno signal yet
22:50 NYoffonno signal yet
Trade geometry, as the messages describe it
Reward multiples, scale-out1:1 · 2:1 · 3:1, thirdsdeclaredawaiting a real signal ⚠ hand-fired tests are deliberately not counted — their geometry is ours, not the strategy's
Strategy inputs changed from your defaults
Max Trades Per Day74declared
Enable Webhook Alerts?falsetruedeclared
Ticker SymbolNQM2026US100declared
Execution limits (ours, not yours)
Risk per trade0.5 % of balancedeclared
Position cap5.0 lotsdeclared
Daily loss limit3.0 %declared
Concurrent positions1 declared

⭐ Every one of the 66 inputs in your strategy, with the 13 the pilot changed marked, is on its own page: the full settings table → It is read straight out of your Pine source, so the defaults column cannot drift — and it is the proof that nothing else was touched.

Assessment

What this data does and does not support

Each finding recomputes from the records on every rebuild, heading included. ⚠ The reasoning behind them does not — it updates its numbers but cannot notice something nobody has looked for.

No trade from the strategy itself yet

Every closed trade on record - 0 of them - was hand-fired to prove the plumbing, and all are excluded below. The link from TradingView to this system is the one part that has never been observed carrying a real signal. Until it does, this page reports on machinery, not on a strategy.

·

Every guard has fired at least once, deliberately

0 plans executed, 0 plans refused by the executor, 0 rejected at the relay before the executor saw them. ⭐ Each refusal was a designed stop exercised on purpose during testing — a stale plan, a stop the broker's minimum distance would not accept, a budget already spent, a bad secret. The alternative in each case was trading something other than what the strategy chose. That every one of them fired, and refused rather than adjusted, is the evidence this section can offer today.

·

Next deadline: Vantage demo account, 25 days away

Vantage demo account 2026-10-16 (25d) · nas2m alert 2026-11-19 (59d). ⚠ Each of these stops the pilot with no error, no email and no trace, and a strategy that trades at most four times a day cannot be watched for silence — a lapsed account and a quiet market look identical. ⭐ That is why they are counted down from recorded dates rather than noticed after the fact. The nearest one is the one that matters.

Architecture

How a signal becomes an order

Four hops. Your script owns the first and nothing after it.

1
Your strategy fires
An alert on order fills sends the JSON your Pine already builds — direction, size, stop and all three targets. TradingView posts it from its own servers.
2
A relay catches it
A small program on a free Cloudflare account. It exists because TradingView needs a public address to post to and a home PC does not have one. It stamps the message, refuses duplicates, and holds it.
3
The executor collects it
An expert advisor inside MetaTrader asks the relay for anything new. ⭐ It reaches out, so nothing needs to reach in — no port forwarding, no fixed address, nothing exposed.
4
The broker fills it
The stop goes to the broker with the order, so a dead executor still leaves a protected position. Thirds close at each target and the stop moves to breakeven on its own.

Running it without us

⭐ The relay is yours to own, not ours to rent you. Everything on this page runs on infrastructure you would control.

The relay is a free Cloudflare account and four commands. Your account, your URL, your secrets. Nothing runs on our side, and nothing breaks if we disappear.

The executor is a single file. It drops into MetaTrader's Experts folder and attaches to a chart. Its risk settings are a text file it re-reads every thirty seconds, so they change without a dialog and without a restart.

The alert message is a wrapper around your own. ⭐ That is why your Pine needs no changes at all — TradingView substitutes your strategy's JSON inside it.

None of this is hypothetical — the pack is built and waiting: the compiled executor (123 KB), a settings template, a preset template and a setup guide written from what actually went wrong. ⚠ It contains no credentials of any kind; the URL, token and account number are yours to fill in. Ask and it is yours.

Status

What this run has and hasn't shown

Proven, on real orders
  • The whole plan travels in the entry message — every later message is ignored, so none can be lost, duplicated or misread.
  • The stop is attached to the order, and the final target parked at the broker.
  • A position stopped out by the market was reported as a loss — which your strategy's own message stream would have reported as a take-profit.
  • A stale plan is refused rather than traded late.
  • The executor was killed mid-trade and recovered its position, cursor and settings.
Not yet observed
  • A signal arriving from TradingView. Every plan so far was hand-fired. This is the one link in the chain that has never carried a real message.
  • Breakeven succeeding — it needs price to move off the entry far enough to clear the broker's minimum stop distance. ⭐ The executor started reporting this on 16 September, so the first success will be recorded rather than missed.
  • Any target filling. The one real trade stopped out before its first target, so partial closes and breakeven have never run on a strategy signal.
  • Anything at all about whether the strategy makes money. See the assessment.