Where Expert Advisor Track Records Meet Quantitative Truth

A 500% return with 80% drawdown looks better than 100% return with 15% drawdown.. but which would you actually trust with your money?

View Leaderboard

Sortino

Calmar

Ulcer Index

Neg. Skewness

Autocorrelation

Float Stress

The problem

Why simple performance metrics are not enough

A strong equity curve can hide serious structural risk. AlgoScore reveals what headline numbers miss.

My Holy Grail EA

Live since Jan 2023

Verified

Gain

+342%

Drawdown

67%

Win Rate

84%

Equity Curve

Looks impressive at first glance.

ALGOSCORE

0/100

Poor

Persistent drawdowns

Tail risk exposure

Floating losses

Return smoothing

This is why AlgoScore analyzes what headline metrics miss.

The Solution

One score that tells you what returns alone never will

AlgoScore analyzes six deeper dimensions behind every strategy and compresses them into one score you can understand instantly.

0Poor

Steady Growth Pro

by DemoTrader

Trend / Breakout4 pairsSince Jan 2023

Gain

+186%

Drawdown

14%

Monthly

3.1%

2.4

Win Rate

71%

Trades

2,841

Performance

Risk

Consistency

Stress

Metric Breakdown

Sortino Ratio

2.14

Weight: 30%

Calmar Ratio

1.62

Weight: 20%

Ulcer Index

3.41

Weight: 20%

Neg. Skewness

-0.82

Weight: 10%

Autocorrelation

0.12

Weight: 10%

Float Stress

0.08

Weight: 10%

6 risk dimensions0-100 score rangeUpdated daily

Methodology

How it works

From raw data to final score in five transparent steps.

Data Collection

Daily returns and account statistics pulled from verified Myfxbook track records via API every day at 01:00 UTC.

Metric Computation

Six risk-adjusted metrics computed across two time windows: all-time and recent (365 days).

Statistical Testing

Ljung-Box significance testing on autocorrelation. Penalties applied only when statistically confirmed (p < 0.05).

Confidence Adjustment

Bayesian-inspired shrinkage toward neutral for short records. Longer proven history rewarded with diminishing returns. This produces the v1.2 engine score.

Manual Overlay & Calibration

Up to three transparent multipliers — cash-flow/true-TWR, grid/martingale and scalability — correct for risks automation and Myfxbook can't see. The result is then mapped onto a fixed, percentile-anchored display scale (never changing the ranking) to produce the published score.

The Scoring System

What is behind the score

Each metric targets a different weakness that simple returns can't reveal.

Performance

Sortino Ratio

Measures return relative to downside deviation only. Unlike the Sharpe ratio, upside volatility is not penalized, making it ideal for trend-following or momentum strategies.

Calmar Ratio

Annualized return divided by maximum drawdown. Answers a direct question: how much return does this strategy generate per unit of worst-case loss?

Risk

Ulcer Index

Captures both the depth and duration of drawdowns from the equity peak. A strategy that stays underwater for months scores worse than one that recovers quickly.

Neg. Skewness

Detects left-tail fat in the return distribution. Strategies with occasional large losses hidden behind steady small gains will show elevated negative skewness.

Consistency

Autocorrelation

Tests for serial dependency in weekly returns using the Ljung-Box statistic. High autocorrelation can indicate grid trading, martingale layering, or artificial equity smoothing.

Stress

Float Stress

Measures persistent floating losses relative to account balance. Catches strategies that hold losing positions open for extended periods to avoid realizing losses on the equity curve.

Manual Overlay

The six metrics produce the automated v1.2 engine score. On top of it, up to three transparent multipliers correct for risks that automation and Myfxbook can't see — raw score = engine score × cash-flow × grid/martingale × scalability. Manual multipliers are only ever 1.00 or a value between 0.50 and 0.80, and every applied penalty is shown with its reason on each strategy's detail page. Finally, the raw score is expressed on a calibrated display scale anchored to the retail EA population — a fixed, strictly increasing mapping that makes the numbers intuitive (a median retail EA reads about 70) without ever changing how strategies rank.

Cash-flow / True-TWR×0.00 - 1.00Transcribed from an external true time-weighted-return service that strips out deposits and withdrawals made during drawdowns, exposing equity curves that were smoothed by balance changes.

Grid / Martingale×0.50 - 0.80A residual penalty for grid, martingale and averaging risk that the automated metrics and Myfxbook do not fully capture. Heavier when the true drawdown is understated.

Scalability×0.50 - 0.80A penalty for strategies unlikely to scale to larger accounts: night scalping, ultra-short holding times, spread or latency sensitivity, or only small-account evidence.

Want the full methodology? Read the AlgoScore White Paper

Weighting System

How the score is weighted

A single engine score from four weighted components, before the manual overlay is applied.

Performance

50%

SortinoCalmar

Risk

30%

Ulcer IndexSkewness

Consistency

10%

Autocorrelation

Stress

10%

Float Stress

Score Ranges

How to interpret the score

AlgoScore ranges indicate the overall quality of a strategy.

80-100ExcellentStrong risk-adjusted returns, controlled drawdowns, and consistent behavior across all dimensions.

60-79GoodSolid fundamentals with minor weaknesses. A good candidate for further due diligence.

40-59FairMixed signals. One or more metrics indicate elevated risk or inconsistency.

0-39PoorSignificant concerns in multiple areas. High drawdowns, tail risk, or signs of curve-fitting.

FAQ

Frequently Asked Questions

How often are scores updated?

Every day at 01:00 UTC. The scoring engine pulls the latest data from Myfxbook and recomputes all six metrics for every Expert Advisor automatically.

Can a short track record get a high score?

It's harder by design. The confidence layer shrinks scores toward 50 for short records, so a newer track needs more time to prove itself. But the score is driven purely by statistical quality, not by duration bonuses.

Why not just use Sharpe ratio?

Sharpe penalizes all volatility equally -- including upside gains. For trend-following or momentum strategies, this is counterproductive. Sortino ratio (which we use instead) only penalizes downside deviation.

What does the autocorrelation penalty detect?

It flags strategies that may be artificially smoothing their equity curve -- for example through position averaging, grid trading, or correlated entry layering. The penalty is only applied when a Ljung-Box statistical test confirms significance at the 5% level.

Why are some EAs' published scores lower than their engine score?

Because of the manual overlay. The automated v1.2 engine can only see what Myfxbook reports, but Myfxbook sometimes understates the true drawdown, balance changes can smooth an equity curve during a drawdown, and some strategies don't scale to larger accounts. The overlay applies up to three multipliers (each transcribed from a more accurate source or a documented manual judgement) to correct for this. Every applied penalty is shown with its reason on the strategy detail page, and we apply it to our own EAs as readily as to others.

What is the calibrated display scale?

The raw score (engine score times the overlay multipliers) is mapped onto a display scale anchored to percentiles of the retail EA population, so the published numbers are intuitive: a median retail EA reads about 70, and the strongest systems read in the high 80s and low 90s. The mapping is a fixed, strictly increasing curve, so it can never change how two strategies rank against each other, and it is frozen: your score never moves because some other EA changed. The full raw-to-display breakdown is shown on every strategy's detail page.

Is AlgoScore investment advice?

No. AlgoScore is a quantitative analysis tool designed to facilitate comparison and due diligence. A high score does not guarantee future performance.

Where does the data come from?

All performance data is sourced from verified Myfxbook track records. Myfxbook provides time-weighted returns that account for deposits and withdrawals. Where Myfxbook is known to be inaccurate, the manual overlay corrects for it using more accurate sources such as MQL signal drawdown reports.

Ready to see the scores?

Explore the live leaderboard and see how every EasyAlgos Expert Advisor performs under rigorous quantitative analysis.

View Leaderboard

Read the White Paper