Strategy Spotlight: How We Evaluate and Share Real Trading Results

Most trading “results” you find online are carefully curated fiction. A vendor runs 10,000 optimization passes, picks the one equity curve that goes up and to the right, slaps it on a landing page, and charges $299 for a signal subscription. The forward test never happened. The account blew up in three weeks.

This is the first post in an ongoing series we are calling Strategy Spotlight. We take a trading strategy, run it through a rigorous evaluation process, forward test it on real tick data, and publish everything. The good, the bad, and the ugly equity curves. No cherry-picking. No hiding the drawdowns.

We are a VPS hosting company. We make money when traders keep trading, not when they buy a magic indicator. Our incentive is aligned with yours: traders who run strategies that survive stay on our servers for years. Traders who blow up cancel after a month.

Why Transparency Matters More Than Ever

The algo trading space has a credibility problem. Open any trading forum and you will find hundreds of threads showing backtests with profit factors above 4.0, Sharpe ratios above 3.0, and drawdowns under 5%. What you will not find is the follow-up post six months later showing what actually happened in live markets. Those threads go quiet. The poster disappears.

We host the infrastructure that runs these strategies, and we watch the patterns repeat. The strategies that look modest in backtest tend to survive. The ones that look like money printers tend to implode. This series exists to close the gap between simulation and reality. Every strategy we spotlight will include:

The logic behind the strategy in plain language
Backtest results with full parameter disclosure
Forward test results with real spreads and slippage
An honest assessment of what worked and what did not

We are not selling these strategies. We are documenting what real systematic trading looks like.

Our Evaluation Methodology

Not every strategy earns a spotlight. Here is how we decide what to feature and how we measure it.

Selection Criteria

We look for strategies that meet three conditions:

The logic is explainable. If we cannot articulate why the strategy should work in plain English, it is probably curve-fit noise. Market structure, not pattern-matching on historical accidents.
The backtest is honest. Walk-forward tested on out-of-sample data. Realistic spread and commission assumptions. If the strategy only works with zero-spread backtesting, it does not work.
It trades frequently enough to evaluate. A strategy that takes three trades per month needs years before you can say anything meaningful. We need enough trades to reach our minimum sample size within a reasonable forward test window.

What We Measure

We track the same core metrics from our Strategy Performance Metrics guide:

Expectancy per trade — the average dollar amount you expect to make or lose on each trade. If it is negative, nothing else matters.
Profit factor — gross profit divided by gross loss. Below 1.2, execution costs will likely eat your edge. Above 3.0, be suspicious.
Sharpe ratio — risk-adjusted return per unit of volatility. Above 1.0 is acceptable. Above 2.0 is strong.
Maximum drawdown — the worst peak-to-trough decline. The number that determines whether you can psychologically survive the strategy.
Trade count — thirty trades is the bare minimum to draw any conclusions. We prefer 100 or more before we publish.

Forward Test Requirements

Every spotlight includes forward test data. Our minimum bar:

At least 30 completed trades (we aim for more, but this is the floor)
Executed on real tick data with actual market spreads
Run on our own VPS infrastructure under the same conditions our clients experience
No parameter changes during the test — settings are locked before the first trade

If a strategy looks great in backtest but falls apart forward, we say so. That is the entire point.

Strategy Spotlight #1: Opening Range Breakout on Indices

For our inaugural spotlight, we chose a strategy that has been around for decades and is well understood: the Opening Range Breakout (ORB).

The Concept

During the first 15 to 30 minutes of a session, price establishes a range as the market digests overnight information. Once price breaks convincingly above or below that range, it tends to continue as momentum traders and algorithms pile in.

This is a structural edge rooted in how markets open. Institutional order flow concentrates around the open, and the range reflects the initial battle between buyers and sellers. When one side wins decisively, the follow-through is tradeable.

Parameters

Instruments: DAX (DE40), DOW (US30), NAS100 (US100)
Timeframe: M5 (5-minute bars)
Range period: First 30 minutes of the European session (DAX) and US session (DOW, NAS100)
Entry: Break above range high (long) or below range low (short), confirmed by the M5 candle closing beyond the level
Stop loss: Opposite side of the range, with a minimum of 9 points
Take profit: 2.0x the stop distance (target of 18 points when stop is 9 points)
Max one trade per session per instrument
No trading on major news days (FOMC, NFP, ECB rate decisions)

Forward Test Results

Metric	DAX	DOW	NAS100
Test period	47 trading days	47 trading days	47 trading days
Total trades	38	34	41
Win rate	42.1%	38.2%	44.0%
Average winner	$161	$174	$183
Average loser	$89	$94	$97
Profit factor	1.31	1.14	1.48
Max drawdown	8.7%	11.2%	7.4%
Expectancy/trade	$14.20	$7.60	$22.10
Sharpe ratio (annualized)	1.08	0.71	1.34

Combined across all three instruments: 113 trades, 41.6% win rate, profit factor 1.31, expectancy $14.80 per trade, max portfolio drawdown 9.3%.

These are not numbers you put on a marketing brochure. A profit factor of 1.31 and a 42% win rate will never sell subscriptions. But this is what a modestly profitable, survivable strategy actually looks like. The expectancy is positive. The drawdowns are manageable. The trade count is approaching statistical relevance.

Is it going to make you rich? No. Is it better than most of what we see deployed on our servers? Honestly, yes.

Lessons Learned

Every forward test teaches you something the backtest did not. Here is what stood out.

Slippage was worse than expected on breakout entries. Breakout strategies by definition enter when momentum is accelerating, which means you are competing with every other breakout system for the same fills. Our average slippage was 1.2 points on DAX entries versus the 0.5 points we assumed in backtesting. On a 9-point stop, that extra 0.7 points shaved roughly 15% off the backtest expectancy.

DOW underperformed significantly. The backtest showed DOW performing comparably to the other two indices. In forward testing, it was noticeably weaker — likely because the choppy, range-bound regime on DOW during the test period is unfavorable for breakout logic. A useful reminder that regime sensitivity does not show up in a single backtest window.

News day filters saved us at least twice. We skipped seven trading days due to scheduled high-impact events. On at least two of those days, the opening range was violated in both directions within minutes — a full stop loss hit on a false breakout. Simple filters like this are boring. They also work.

The 2:1 reward-to-risk ratio held up. Winners averaged almost exactly 2x the size of losers across all three instruments. When the breakout works, it works properly. When it fails, the stop catches it quickly. The asymmetry is real.

Important Disclaimers

Everything in this series is for educational purposes only. We are not financial advisors and we are not recommending you trade any strategy featured here. Past performance — whether backtested or forward tested — does not guarantee future results. Markets change. Edges decay.

Forty-seven trading days is a start, not a conclusion. A strategy needs months of live data before you can have real confidence. We will continue tracking this one and update results as the sample grows.

If you trade any strategy, do so with capital you can afford to lose and sizing that lets you survive the worst-case drawdown.

What’s Next

This is the first in a series. We plan to spotlight mean reversion, trend following, session-based, and multi-timeframe strategies across different instruments and market conditions.

If you are running a strategy on your FXVPS server and want to share your results (anonymized or otherwise), we would genuinely like to hear from you. The more real data points this community has, the harder it gets for the snake oil sellers to operate.

Next up: a mean-reversion approach on forex majors — higher win rates than today’s breakout strategy, but a completely different risk signature. Stay tuned.

This is part of our ongoing library of practical guides for algorithmic traders. For the metrics framework referenced in this post, see Strategy Performance Metrics That Actually Matter. For the full pipeline from backtest to live, see Backtesting to Live: The Complete Pipeline.