The Sounding · the week’s read
I read a building swell on gold this week.
80% weekly range · confidence cone
How to use this: the fan is the water’s width, not a course. A wider fan is more room a stop needs to breathe; traders carry less sail when the swell builds. The numbers are odds, never a call.
Count the odds
Touch odds
How often a week like this reaches a mark. Odds it tags, not a promise it holds.
Sharpness
tightnormalwide
The quant method
Step 1 of 5. You have the water. Now weigh the trade you are eyeing.
The Reckoning · your own numbers
How much sail can you carry before a bad run sinks you?
$
$
Risk of ruin · survival over 100 trades
Blow-up odds · 100 trades
about 0
Position size
0
Rough-week loss on this size · expected shortfall
$0
Money at risk · this trade
$0
Stop vs the week’s normal floor
ok
How to use this: put in your own account and stop, then drag the risk. Watch the survival curve: at 1 to 2% a losing streak can’t wipe you; at 5% and up the deep ends chew you out over a hundred trades. The stop-vs-floor check flags amber when your stop sits inside the week’s normal breathing room, where normal swell shakes it out.
The quant method
Your real bets · are these independent?
Effective number of bets (Meucci)
2.9
Margin runway · leverage as a vol multiplier
Your account, stop and risk stay on your device. Nothing is typed here is sent or stored. The ruin, ES, correlation and margin math run in your browser off the banked model output.
Step 2 of 5. You have sized it. Now see which playbook this week’s water rewards.
The Routes · the regime lens
This week reads ranging water. Here is what each route weighs in it.
How to use this: pick a style. You get its own step-by-step plan on THIS week’s water: the levels it watches, the conditions it waits for, the odds it even gets a setup, the stop it implies and the rough-week cost if it is wrong. Taught the same across every style so you compare and choose. Never a course to sail.
The quant method
Step 3 of 5. You know the playbooks. Now see what could blindside you.
The Storm · event risk
One squall on the calendar this week.
Next scheduled release
US Non-Farm Payrolls
How much a print like this moves gold
How to use this: know when the water gets rough and by how much a print like this has moved the asset before. Traders often carry less sail into a release. It is history, not a forecast of direction.
The quant method
Weekend / gap gauge
A stop is a trigger, not a guaranteed price. Over a weekend, news prices in with no path between. Here is how far this asset has gapped.
Preview · sample data
Forward implied volatility · term structure
Step 4 of 5. You see the storm. Now check the read is honest.
The Log · the receipt
Here is my whole log, hits and misses.
Cone coverage, out of sample · claimed 80%
Backtest
Liveweekly forecasts · still resolving
Too few to trust the number yet. Shown anyway, because hiding the small sample is how a scam starts.
held
broke out
still open
How to use this: this is the calibration record a signal-seller can never show you. When the band claims 80%, did price stay inside about 80% of the time? Backtest and live are labelled apart; low calibration is meant to make you widen your own margin, not narrow it. Drift is monitored, never a trade.
The quant method
Step 5 of 5. You have read the water, weighed the trade, seen the playbooks, the storm and the honesty of the read. You now see the whole picture. You set the course.
The Reality Cone · what to expect
Put in R500, thought you’d double it in a month? Here is the honest fan.
Equity over 100 trades · a modest 52% edge at this risk
Where a typical path lands
1.00x
Odds you double it
0%
Odds you are down 30% or worse at the end
0%
How to use this: even a real, winning edge crawls, and even winning paths draw down hard along the way. This is what a modest honest edge looks like, so the month-one double is measured against reality. Drag the risk and watch the median stay flat while the odds of a deep drawdown climb.
The quant method
The fan is a Monte-Carlo simulation of a 52% edge, 1:1 payoff, 100 trades, banked in the model. It is not your account and not a projection of your returns.