Algorithm Decay: When Your Profitable EA Stops Working

Your Profitable EA Is Slowly Dying—And You Haven't Noticed Yet

The traders who make consistent money don't hold the same strategy for three years unchanged. And they don't abandon it either. They rebuild it.

Here's what happens: your EA runs great for 6 months. You're hitting 65% win rate, 2.5:1 risk/reward. You stop touching it. Then month 7 comes. Win rate dips to 58%. You think it's variance. Month 8: 53%. Month 12: 38%. By the time you notice—usually after a losing streak costs you $5K—the strategy is half as profitable as it was. That's algorithm decay.

It's not your fault. It's not your EA's fault. It's how markets work.

Three Kinds of Decay Kill Profitable Strategies

Algorithm decay isn't one thing. It's three.

1. Parameter drift. Your EA optimizes for a specific market regime. When volatility changes, when the Fed raises rates, when institutions rotate sectors—the parameters that worked stop working. A moving average crossover that crushed in low-volatility 2022 gets whipsawed in the chop of 2024. The code didn't break. The market it was tuned for disappeared.

2. Regime change. Profitable strategies often exploit specific market behavior. When mean reversion shifts to momentum, or volatility compression turns to expansion, your EA still runs. It just loses. Regime changes are silent. You don't get a notification. You get smaller and smaller profits until there are none.

3. Competition. If your strategy works, so does someone else's. The more people who run the same signal, the faster the edge decays. A breakout strategy that made consistent 2% monthly five years ago might now make 0.3% before slippage, because every retail trader with an EA is trading the same breakout. The edge didn't disappear. It got crowded.

Most traders experience all three at once. They don't know which one is killing their strategy. So they can't fix it.

How Decay Stays Hidden Until It's Expensive

Algorithm decay is the silent killer because it looks like variance.

A profitable strategy with 65% win rate will have losing months. You expect that. So when month 2 underperforms, you don't panic. Month 3 too? Still variance. By month 5, you're frustrated. By month 7, you're done. You assume the strategy never worked, or the market changed, or you're unlucky. You shut the EA off and try something new.

But here's the cost: every month of decay costs you the gap between what you made last year (2% monthly) and what you're making now (maybe 0.5%). Over 12 months, that's 18% in lost performance. On a $50K account, that's $9,000 in opportunity cost. And you didn't even know the strategy was decaying.

The traders who catch decay early—within 2-3 months—lose $1K-$2K. They fix it. Done. The traders who ignore it until it's obviously broken lose $5K-$15K before they act. That's the difference between catching drift and letting it compound.

How Professionals Monitor and Rebuild

Professional traders don't set their EAs and forget them. They monitor four metrics obsessively:

• Monthly win rate vs. historical average. If it drops more than 5-10 percentage points, decay is happening. Time to dig.
• Consecutive losing trades. Even with a 60% win rate, you expect losing streaks. But if they're getting longer (3 losses becomes 5, then 7), your parameters are drifting.
• Average profit per trade. This is the killer metric. If your average profit per winning trade shrank from $200 to $80, you have a decayed strategy. The signals are worse.
• Slippage and spread creep. Sometimes decay is just cost inflation. Real trading slips more than backtests predict. If slippage doubled, profitability gets cut in half.

Once they spot decay, professionals rebuild. One of three moves:

• Reoptimize the parameters to the last 6 months of data. This works when the core logic is sound but the regime shifted.
• Combine with a second strategy that captures a different edge. One catches momentum, another catches reversals. Together they're more stable.
• Rebuild from scratch on fresh data, keeping the core logic but letting the market tell you what parameters matter now. This is hardest but often most profitable.

A single reoptimization takes a few hours. A combination takes a day. A rebuild takes... well, it depends. Most traders either can't do it or it takes weeks.

DIY Decay Detection vs. Outsourced Monitoring

You can monitor decay yourself. Check monthly, log the metrics, spot the trend. Most traders don't because it's boring. So decay compounds until losses force their hand.

Or build monitoring into your EA. A dashboard that shows win rate, average profit, consecutive losses, and a health score that alerts when metrics drift. That requires someone who knows MT5 API, can pull clean trade data, and automate without errors. Alorny builds custom EAs with built-in monitoring dashboards. Every EA includes performance tracking, decay detection, and reoptimization recommendations. Monthly reports show exactly what your strategy is doing and when it needs rebuilding. Starting from $300 for a simple custom EA, you get the monitoring included. Higher complexity (ICT, SMC, multi-timeframe strategies) starts at $500+.

The cost difference between catching decay in month 2 ($300 rebuild) versus month 7 ($9,000 in opportunity cost) is obvious. But most traders don't see the problem until month 7.

The Real Cost of Algorithm Decay: 12 Months of Opportunity Loss

Let's be specific.

Scenario: You built a profitable EA 12 months ago. It ran at 2% monthly for 6 months. Then it decayed. Months 7-12 it ran at 0.5% monthly. You didn't notice until month 10 because the losses were gradual.

Your account: $50,000

With the original strategy (2% monthly): $50K → $56,127 after 12 months.
With decay (2% for 6 months, 0.5% for 6 months): $50K → $53,062 after 12 months.
Opportunity loss: $3,065 in 12 months.

Now add the time cost. You spent 40+ hours wondering why the EA wasn't working, backtesting new ideas, rebuilding twice from scratch. At $100/hour, that's $4,000 in lost time.

Total cost of decay: $7,065 in real dollars and lost time.

A $300 custom EA with built-in monitoring would have caught the decay in month 2. A one-week reoptimization costs $400. Total cost to fix: $700. Savings: $6,365.

This isn't theoretical. Every trader who has run more than one strategy knows this math.

What to Do Now: Diagnose Your Current EA

If you have an EA running right now, here's the diagnostic:

1. Pull the last 3 months of trade history from your MT5 terminal.
2. Calculate: win rate, average profit per winning trade, consecutive losing trades, average trade duration.
3. Compare to 6-12 months ago. Did win rate drop? Did average profit shrink? Did losing streaks get longer?
4. If all three are worse, you have decay. It's not variance. It's not luck. It's decay.
5. Next step: either rebuild the EA yourself (which takes time) or get help rebuilding it (which takes a week and costs $300-$500).

Most traders don't do this diagnostic. They assume if it's not obviously broken, it's fine. It's not. Silent decay is the most expensive mistake because you don't feel the cost until it's too late.

Alorny can audit your current EA for decay signals and rebuild it to current market conditions. Send your EA code and 6 months of trade history. We'll diagnose exactly what's decaying and rebuild it. Price: from $400 to $600 depending on complexity. Includes full backtest report and 30 days of revisions.

Key Takeaways

• Algorithm decay is silent. Your profitable strategy doesn't fail overnight. It degrades so slowly you mistake it for normal variance. By the time you notice, you've lost thousands.
• Three types of decay kill strategies: parameter drift (your settings no longer fit the market), regime change (the market behavior your strategy exploited disappeared), and competition (your edge got crowded out).
• Professionals catch decay in 2-3 months. They monitor win rate, average profit per trade, losing streaks, and slippage. When any metric drops 5-10%, they diagnose and rebuild.
• DIY monitoring is tedious. Outsourcing is cheap. A built-in monitoring dashboard costs $300 one-time. Ignoring decay until it destroys your strategy costs $5K-$15K.
• Rebuild, don't restart. When decay hits, reoptimize the parameters, add a complementary strategy, or rebuild with fresh market data. The core logic usually works—it just needs tuning.

Here's what winning traders do: they rebuild before the strategy breaks, not after. They don't wait for losing streaks. They monitor monthly. When metrics drift, they act. That's the difference between traders who make 2% monthly consistently, and traders who made 2% last year and can't figure out what went wrong.

Your EA is either performing at peak or starting to decay. There is no middle. Time to check which one.