F FunTrades

precision · patience · profit

The MIDAS Model

A precise, repeatable way to trade the gold discount, and the premium. Entry, stop, and target, all decided before I click.

The MIDAS trade, played out Price arrives from the left and taps the limit entry at 0.75 (stop at 0.9), then runs down through the reward zone to take profit at 0.5. 1.0 0.9 STOP 0.75 ENTRY 0.6 TP · 1:1 0.5 TP · 1:1.6 0 the tap

MIDAS is a scalping model I trade on gold futures (GC / MGC), on the 1- to 5-minute charts (mostly the 2-minute), built from three things, read together.

Price

The impulse leg, and the structure around it.

Fibonacci retracement

Entry, stop, and target at fixed levels.

Volume profile

The value area, as confluence for the entry.

It's the model behind my first prop-firm payouts.

The mechanics

mechanical · no judgment

Find a strong price leg in one direction, draw the fib across it, and place the orders at fixed levels, the same way every time:

  • Entry · limit Fills only on a tap back into the level, no chasing. 0.75
  • Stop Fixed risk: one clean R, decided before the click. 0.90
  • Take-profit 0.6 for a clean 1:1; 0.5 for 1:1.6 when the bounce is strong. 0.60 / 0.50

The entry is a limit order at 0.75: it only fills if price returns there. Stop sits at 0.9. The default target is 0.6 for a clean 1:1; I'll move it to 0.5 for a 1:1.6 when the bounce is clean and there's liquidity above (or below) to support holding a runner.

FIG. 01 · MECHANISM

The shape I'm looking for: liquidity gets swept (sometimes more than once), the leg forms, and price taps back into the 0.75 entry before continuing. This shows the cleanest version: value area sitting exactly at 0.75. Anywhere between that level and the POC still counts, just a notch behind the best case.

What I actually check

the judgment part
Filter 01

Leg size

The leg has to be meaningful relative to the day's range. A small leg on a big-range day gets skipped: if the move is only a small fraction of the full daily range, it's too weak to build a trade on.

Filter 02

Value-area confluence

Best case: the developing VAL (for longs) or VAH (for shorts) sits right at the 0.75 entry. Still valid: anywhere between that level and the POC. Outside the value area entirely (or sitting right at the POC alone) is weaker, and usually a skip.

Filter 03

Liquidity

It helps when a recent high or low gets swept just before the leg that builds the fib forms. That sweep is often what kicks the real move loose.

Filter 04

Regime

This setup performs better in ranging conditions. When price is trending strongly outside the value area, skip it.

Filter 05

Timing

No trades before 3:00 AM New York time. The volume profile needs time to develop before the VAL, VAH and POC levels mean anything.

FIG. 02 · LIVE EXAMPLES

Real trades. Each is the same shape: liquidity swept, the leg forms, and price taps back into the 0.75 entry, across the 1-, 2- and 5-minute charts. Click any one to see it full-size.

By the numbers

backtested + live · one sample
Setups
363
Win rate · 1:1
71.3%
Profit factor
2.49
Expectancy
+0.43R
Total
+155R

363 MIDAS setups on gold, backtested and live combined, each scored mechanically at the fixed 1:1 take-profit (idealised R, before fees and slippage). Letting winners run to the 1:1.6 target instead: 58.3% win rate, +0.52R per trade.

One leg. One fib. One filter. Gold only.

A note on risk. This describes a personal trading process and is shared for educational purposes only. It is not financial advice. Trading futures involves substantial risk of loss and is not suitable for everyone. Past results do not guarantee future outcomes.