NQ Alpha Engine is a fully automated, model-driven strategy trading Micro E-mini Nasdaq 100 futures. No human override. No emotional interference. Rules execute every time.
Join Telegram to receive live trade alerts and daily briefings.
| Year | Jan | Feb | Mar | Apr | May | Jun | Jul | Aug | Sep | Oct | Nov | Dec | YTD |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2026 | — | — | +36.6% | +25.6% | +1.0% | +28.4% | — | — | — | — | — | — | +122.6% |
Includes estimated broker commissions and strategy subscription fees.
NQ Alpha Engine is built around a core idea: certain price levels on the Nasdaq 100 have a measurable tendency to act as turning points — either support zones where price bounces, or resistance zones where price reverses. The system maps these structural levels daily, then uses an XGBoost machine learning model to score each one based on current market context: volatility regime, trend alignment, momentum, and session conditions. Only levels that clear a dynamic probability threshold generate a trade signal — long or short depending on where price is and what the model expects.
Before any signal reaches execution, it passes through a set of independent regime filters. If the market is in a strong directional trend, a high-volatility expansion, or an otherwise unfavorable environment for mean-reversion entries, the system stands aside entirely. This is intentional — the edge is specific to certain conditions, and forcing trades outside of them destroys it. The model is retrained on a rolling basis so it adapts to shifting market regimes rather than overfitting to historical data.
Risk management is hardcoded into the architecture. Every trade carries a fixed stop loss and a predefined profit target. There is no averaging into losing positions, no manual overrides, and no discretionary exceptions. Once a level has been traded in a session, it is locked out — preventing the system from repeatedly re-entering the same failed setup. Every decision is made by the model, consistently, on every trade.
This is not a signal service. It is a complete systematic trading infrastructure — institutional-grade tooling made accessible at a fraction of the cost to build.
XGBoost ML model trained on 37 market features. Fully automated — no discretion, no emotion, no overrides. Every trade has a defined 200-point stop.
Trades-Own-Strategy certification confirms real capital is at risk on every signal. The strategy manager trades this system live at 100% scaling. Your signal is his trade.
Signals originate from a dedicated server co-located in Chicago — the same data center used by institutional futures traders. Sub-millisecond latency to CME. Trades execute while you sleep.
Every trading morning: ML-generated key levels, pattern fingerprint analysis, scenario model with probability weightings, Fear & Greed index, economic events, regime classification, and overnight context. A standalone research product.
The XGBoost model retrains on a rolling basis to adapt to current market regime. It is not a static backtest. It evolves with the market — a distinction that separates professional systems from retail ones.
Daily content covering trading psychology, behavioral finance, and probability — designed to make you a sharper trader independently of the signals. Strategy Manager is responsive to messages, and shares regular updates so subscribers are informed.
Systematic futures trading is one of the most powerful tools available to individual investors — but it rewards patience and punishes impatience. A few principles will significantly improve your experience.
This system trades overnight and around economic events. Manual execution defeats the edge. AutoTrading through your broker via Collective2 is not optional — it is how the strategy is designed to be followed. Many of the best setups occur outside regular market hours.
Regardless of account size, trade 1 MNQ contract for your first few months. Futures have a different psychological texture than stocks — leverage, overnight gaps, and position management feel different in practice. Build comfort with the system before scaling. The edge compounds over time, not over size.
No systematic strategy wins every trade — including this one. Small losses are a built-in cost of doing business, not a signal that something is wrong. A system with 60% win rate will have runs of 5-7 losses in a row. This is mathematically expected. Judge performance over 50+ trades, not 5. The edge is statistical, not guaranteed on any individual trade.
Academic research (AQR, 2017) recommends allocating up to 20% of investment capital to systematic managed futures — not your entire portfolio. This is not a replacement for diversification. It is a powerful complement to it, with historically low correlation to stocks and bonds. The MNQ micro contract provides precise sizing flexibility for accounts of any size.
The FAQ section covers drawdown expectations, the ML methodology, why there is no multi-year backtest, and how to read the daily briefing. Reading it before your first trade will set the right expectations and make the experience significantly better.
Futures trade 23 hours/day — capturing overnight Nasdaq moves unavailable to stock or ETF traders. They offer true leverage without ETF decay (3× QQQ loses value daily from compounding). They are among the most liquid instruments in the world, with favorable tax treatment in most jurisdictions (60/40 long/short-term capital gains in the US). No asset class combines this level of liquidity, leverage efficiency, and systematic return potential.
This is a fair question, and the honest answer is: backtest data would tell you less than you think, and the live record tells you more.
Nasdaq 100 is highly regime-dependent. The structural dynamics of a trending bull market, a volatility spike, or a macro-driven selloff are fundamentally different — and a model optimized on one regime can look extraordinary in backtesting while failing in the next. Any backtest numbers I could share would be subject to overfitting, look-ahead bias, and the basic problem that the historical market conditions that generated them no longer exist in the same form.
This is precisely why the XGBoost model is retrained on a rolling basis — it is designed to adapt to current regime conditions rather than lock in on a static historical fit. A strategy that retrains regularly cannot produce a meaningful multi-year backtest anyway, because the model itself changes over time.
Current max drawdown of 10% reflects live performance to date. Historically, trend-following systems like this one typically experience max drawdowns of 15–25% over a full market cycle — this is well-documented in the academic literature on managed futures.
The 200-point hard stop per trade limits single-trade risk, but drawdown is ultimately driven by consecutive losses, which can occur in any positive expectancy system. I can't predict future drawdown with certainty — no one can. What I can say is that the system has defined risk per trade, systematic exit rules, and a ratchet stop that locks in profits on winning trades. These are the structural protections against large drawdowns.
For those interested in the research: AQR — "A Century of Evidence on Trend-Following Investing"
TL;DR — Key findings from the study:
NQ Alpha Engine v1 trades a maximum of 3 MNQ contracts. Approximately 90% of all trades are executed at 1 contract — the system's default position size for standard market conditions.
In periods of strong, well-defined trend with elevated conviction, the system may scale to 2 or 3 contracts to maximize participation in the move. This is not arbitrary — position sizing is determined systematically based on current account capital, regime classification, and risk parameters. The account was seeded at $10,000 and has since more than doubled. As capital compounds over time, contract sizing may increase proportionally — or remain conservative to avoid creating meaningful market impact for subscribers scaling at higher multiples.
Any changes to maximum contract size or position sizing methodology will be communicated to subscribers in advance. Transparency on execution parameters is a core commitment of this strategy.
Understanding the psychology behind systematic trading — and why most people fail at it.
Prospect theory, developed by Daniel Kahneman and Amos Tversky, fundamentally changed our understanding of decision-making. Unlike classical economic theories that assume perfect rationality, prospect theory explains how humans actually make decisions under risk — and why those decisions are systematically biased against systematic investing.
We evaluate outcomes as gains or losses relative to a reference point — usually our entry price — not in absolute terms. A trade at breakeven feels like a loss if we were briefly up.
The pain of losing $100 is felt roughly 2–2.5× more intensely than the pleasure of gaining $100. Losses don't just sting — they dominate decision-making in ways that gains never can.
The difference between gaining $100 and $200 feels larger than the difference between gaining $1,100 and $1,200. Early gains feel disproportionately important — causing premature exits.
People overweight small probabilities (why traders fear rare catastrophic losses) and underweight moderate ones (why they dismiss a 62% win rate as "not high enough").
The most dangerous consequence for systematic trading is the disposition effect — the tendency to sell winners too early (locking in the "good feeling") and hold losers too long (avoiding the psychological pain of accepting a loss). When applied to following a systematic strategy, this manifests as unsubscribing after 2–3 losses and re-subscribing after a winning streak — the most reliable way to underperform the strategy's published results.
The algorithm is immune to these biases. It executes rule 200 exactly as it executed rule 1. The subscriber's only job is to let it.
Further reading: Prospect Theory Explained → decodethefuture.org
Futures markets react to events instantaneously. Every economic release, every geopolitical headline, every Fed comment moves the Nasdaq in real time. This is a feature — it creates the inefficiencies the strategy exploits — but for the unprepared observer it can feel like chaos.
Consider this: if residential real estate were as liquid as futures, your home's value would swing violently every time there was a storm, a fire in the neighborhood, or a rise in interest rates. Most homeowners would panic-sell at exactly the wrong moment. The house hasn't changed. The neighborhood hasn't changed. But the ticker would be unbearable to watch. Futures are no different. The underlying edge hasn't changed — only the price at which the market is temporarily willing to offer it.
Ask yourself honestly — do any of these describe you?
These are not character flaws. They are hardwired human responses to uncertainty and loss — the same biases Kahneman and Tversky documented. The solution is not willpower. It is proper position sizing: allocating only capital whose daily fluctuation will not cause anxiety, and committing to a sample size of trades rather than a sample size of days.
Academic research spanning 137 years of market data — including AQR's landmark study on trend-following — suggests that a maximum allocation of 20% of investable portfolio to systematic managed futures strategies has historically produced the best risk-adjusted outcomes. Not because managed futures underperforms at higher allocations, but because it allows the rest of the portfolio to act as psychological ballast.
Someone who has allocated capital they are genuinely comfortable leaving deployed — not money needed next month, not money whose daily fluctuation costs them sleep. Someone committed to a sample size of trades, not a sample size of days. Someone who understands that managed futures, used properly with the right allocation and the right psychology, is one of the most powerful wealth-building tools available to individual investors — institutional in design, now accessible without institutional minimums.
For further research on managed futures as a portfolio complement: AQR — "A Century of Evidence on Trend-Following Investing" →