Ex-Market Maker Shows the Exact VWAP Setup That Gets Traders Funded - Matteo Conti

Finance-focused summary (VWAP-based systematic strategy + funding-challenge math)

Presenter/source: Matteo Conti (ex-market maker; Director of Investments at SQR Capital, a quantitative hedge fund). The video discusses a quantitative, execution–microstructure–driven strategy called “VWAP drift retracement.” It is intended primarily to maximize the chances of passing trader/funding firm challenges (rather than to optimize returns on personal capital).

---

Instruments / markets / tickers mentioned

• Nasdaq 100 futures

  • Ticker/label: NQ
  • Mentioned sizing: 10 micro NQs

• Timeframes / trading windows

  • Uses 5-minute and 15-minute charts
  • VWAP anchored to 9:30 AM Eastern Time
  • Avoids trading in the first hour (9:30–10:30 EST)
  • Trading stops after 3:30 PM ET
  • Positions closed by 3:55 PM ET

---

Strategy: “VWAP drift retracement” (systematic entry framework)

Core idea (microstructure rationale)

• In banks/brokerages, it is stated that 90–95% of execution orders are executed using VWAP execution algorithms (i.e., execution tends to cluster around VWAP).
• When price retraces toward VWAP, execution activity intensifies.
  • If selling algorithms dominate, price can show a bearish impulse
  • If buying algorithms dominate, price can show an upward impulse
• Therefore, the strategy:
  • identifies the day’s drift/trend via VWAP behavior
  • trades the first pullback toward VWAP (the “drift retracement” / “recoil”)

Explicit methodology / step-by-step rules (as described)

1. Chart inputs

   • Load 5-minute NQ futures chart
   • Load 15-minute NQ futures chart
   • Compute VWAP from the 15-minute chart, then apply it onto the 5-minute chart

2. Define directional “drift” (15-minute VWAP trend)

   • Every 15 minutes, check whether a drift/trend condition exists
   • (Example long side given first)

3. Long condition (trend = bullish drift)

   • Price is above VWAP
   • VWAP is rising over the last 15 minutes (green state)
   • Over the last hour, NQ increased by at least +0.1%

4. Short condition (trend = bearish drift)

   • Price is below VWAP
   • VWAP is falling over the last 15 minutes
   • Over the last hour, NQ decreased by at least -0.1%

5. No-trade window

   • Do not trade between 9:30 and 10:30 EST

6. Trigger (entry timing)

   • After trend conditions are met and trading is allowed:
     • Long trigger: open on the first red candle pulling toward VWAP
     • Short trigger: open on the first green candle after trend conditions are met

7. Order mechanics

   • Entry is at the opening of the next candle
   • Then the strategy appears to send a market order when the candle closes (The execution phrasing is noted as somewhat inconsistent in the subtitles, but the intended meaning is: mechanical next-candle open, confirmed at close.)

8. Add-on behavior

   • The strategy attempts to enter on the first meaningful pullback toward VWAP, even if the pullback is close—as long as the trigger occurs.

---

Risk management / exits (explicit numbers + safeguards)

Exit targets / stops (points)

• Long positions

  • Risk: 80 points
  • Gain target: 40 points
  • Described as negative reward-to-risk

• Short positions

  • Risk: 80 points
  • Gain target: 50 points
  • Also described as negative reward-to-risk

Because funding challenges often emphasize consistency passing rules, the approach prioritizes pass probability over high payoff ratios.

Safeguards / trading limits

• One position at a time
• Maximum 4 operations per day
• Maximum 2 losing trades per day (“2 leaks”)
  • If you get two consecutive day losses, stop trading for the day
• No new trades after 3:30 PM ET
• Close all positions at 3:55 PM ET

---

Backtest / validation design (anti-overfitting framing)

Data scope and out-of-sample claim

• Strategy developed using data from 2020 to 2024
• Claim: used operationally for the last six months to pass funding challenges (per guest statement)
• Performance described as:
  • “equity since 2021”
  • from 2024 to “yesterday, August 2, 2026” treated as out-of-sample

Overfitting / in-sample vs out-of-sample explanation

• Overfitting is described as tuning too many rules/parameters to fit past data.
• Method described:
  • Split data (example given): research/tune on the first half
  • Freeze parameters, then apply them to the second half
• For this strategy:
  • Inputs optimized up to 2024
  • Then applied through 2024–2025 with “excellent” performance

---

Performance metrics & explicit probabilities (funding-challenge “pass” math)

Strategy-level return and win/loss asymmetry

• Stated rate of return: 64%
• Reported win/loss-related metrics (as described):
  • “profitability rate” of 65%
  • “Average win rate” described asymmetrically:
    • 866 when right
    • 1,300 when wrong (Subtitles likely conflate metrics; the key point is asymmetric behavior used in simulations.)

Monte Carlo / simulation for funding-firm test passing

Claims include:

• 20,000 Monte Carlo simulations
• Individual test pass rate: 50% (specifically 49.8%)
• Average time to pass: ~3 days
• With the NQ variant (NQ / 10 micro NQs):
  • approval/pass rate: ~49.8%
  • average time to pass: ~3.4 days
  • within 20-day trading limits

Pass probability across multiple attempts

• 1 attempt: ~49.8%
• 2 attempts: 74.8%
• 3 attempts: 87.3%
• 4 challenges: 93.6% chance of passing at least one

This is branded as the “golden ticket” for fundraising/funding firms.

Cost framing (recommendation through numbers, not direct financial advice)

• Each challenge is said to cost about $50
• Spending up to $200 (4 attempts) is framed as “almost guaranteed” to get financing with 93.6% probability.

---

Disclosures / disclaimers

• The video includes: “Nothing in this video is financial advice.”
• Multiple statements emphasize educational intent and strategy validation concepts.

---

Key cautions / limitations emphasized

• Manual trading is discouraged due to the diligence required; automation is positioned as the way to avoid missing mechanical conditions.
• Even with ~50/50 per challenge, success should not be assumed—pushback is included against hype expectations.
• Urges not to intervene after automation is running:
  • Don’t adjust stops/takes/protocol midstream, which can recreate overfitting in real time.
• Monitoring/risk guidance:
  • Use Monte Carlo to establish expected drawdown distributions
  • If live results exceed expected drawdown ranges, stop the strategy

---

Presenters / sources

• Matteo Conti (guest; ex-market maker; Director of Investments at SQR Capital, quantitative hedge fund)
• Host / interviewer (unnamed in subtitles; repeatedly references Matteo and introduces the segment; also mentions “IQ Capital”)