Why Strategies Decay Over Time: Why Trading Setups Stop Working
Blog by Aliceblue News Desk
06 November 2026
9 minutes
Strategies decay when the small advantage they once had gets weaker. Markets change, more traders start using the same idea, and trading costs eat into thin profits. Traders call this alpha decay. You can manage it by comparing live results with your backtest and deciding in advance when you will pause the strategy.
Table of Contents
Why Do Good Setups Stop Working?
You find a setup that makes money for a few months. Then it starts losing, and you have not changed a single rule.
Strategies decay over time because the market behaviour they depended on shifts, or because too many people begin trading the same idea. A backtest means running your rules on old price data to see how they would have done. This guide shows why decay happens, uses a real Indian example, and gives you a plan to deal with it.
What Is Strategy Decay?
Strategy decay is the slow loss of a trading rule’s ability to make money. Traders call that ability an edge. An edge is a small advantage, such as winning slightly more often than you lose, or making more on wins than you lose on losses.
Every rule rests on a belief about prices. A breakout rule believes that a stock crossing its recent high will keep rising. When that belief holds less often, the rule weakens.
Professionals call the same thing alpha decay, where alpha means the extra return a strategy earns over the market. It affects simple chart setups and automated systems alike. If you trade on fixed rules, expect it at some point.
How Does Strategy Decay Happen?
Decay usually follows a sequence.
- A trader notices a pattern in old prices.
- The rule is adjusted until the past looks good. If it is adjusted too closely, it fits random noise instead of a real pattern. This is called overfitting.
- Costs get ignored or added late. Brokerage, Securities Transaction Tax (STT), GST, stamp duty, and slippage all reduce each trade’s profit. Slippage is the gap between the price you expected and the price you got.
- The strategy goes live and works for a while.
- Other traders find the same idea. When many people enter at the same time and price, fills get worse and the move shrinks.
- Market conditions or exchange rules change.
- Win rate and average profit per trade drift down until the edge no longer covers costs.
Rule changes are a real cause in India. From 20 November 2024, SEBI allowed each exchange to offer weekly index contracts on only one benchmark index, and NSE kept Nifty 50. NSE also moved Nifty weekly expiry from Thursday to Tuesday for contracts expiring on or after 1 September 2025. A backtest that mixes the old and new periods can mislead, so test the recent period on its own.
Practical Example With Numbers
A real Indian case: the 920 Straddle. A straddle here means selling a call option and a put option at the same strike near the current Nifty price. Many traders sold it at 9:20 AM every day. Groww published figures for this strategy over two periods.
1 Jan 2020 to 30 Jul 2024 | After Jul 2024 (about 1.5 years) | |
Total profit | 3,400 points | 673 points |
Profit per trade | about 1.5 points | 0.9 points |
Accuracy (win rate) | about 38% | 36% |
Maximum drawdown | about 387 points | 520 points |
Drawdown is the fall in your capital from a peak. The same rule, with thousands of traders copying it, gave smaller profits and a deeper fall. Groww says the strategy no longer looks profitable once brokerage, STT, and exchange charges are added.
A simple illustration of how costs change the answer. These numbers are made up to show the arithmetic. Expectancy means the average profit or loss per trade:
(win rate × average win) minus (loss rate × average loss) minus cost per trade
Backtest | Live, last 6 months | |
Win rate | 54% | 46% |
Average win | ₹3,200 | ₹2,800 |
Average loss | ₹2,000 | ₹2,100 |
Expectancy before costs | ₹808 | ₹154 |
Assumed cost per trade | ₹300 | ₹300 |
Expectancy after costs | ₹508 profit | ₹146 loss |
The workings are (0.54 × 3,200) minus (0.46 × 2,000) = ₹808, and (0.46 × 2,800) minus (0.54 × 2,100) = ₹154. Over 40 live trades, the loss comes to ₹5,840. The rules stayed the same. Only the results changed.
Key Features of a Decaying Strategy
- It fades slowly over months, which makes it easy to miss.
- The win rate or average win usually drops first.
- Drawdowns get deeper than your backtest showed.
- Small setting changes, such as a 20-day average instead of 17, swing results sharply.
- Costs take a bigger share of every profit.
- Live results stay below the backtest for the same period. Arthalab lists slippage, assumed fills, costs, and market changes as the usual reasons for this gap.
Advantages of Knowing About Decay
Knowing about decay will not make you profitable. It helps you avoid avoidable losses.
- You can stop a weak strategy earlier.
- You can set a loss limit for each strategy before you start.
- You stop treating a backtest as a promise.
- You build better testing habits, such as counting every cost and keeping recent data aside.
- Decisions about when to stop come from numbers, not hope.
Risks and Limitations
- The financial risk is staying in a weak strategy too long. In futures and options (F&O), leverage makes those losses larger. Leverage means holding a bigger position than your own money alone would allow.
- The opposite mistake costs money too. Dropping a good strategy during a normal bad patch means extra brokerage and missed gains.
- Behaviour adds to the problem. After a loss, many traders skip signals, exit early or raise quantity to win the money back. Constant rule changes after each loss lead straight back to overfitting.
- There is also a plain limit. Nobody can say exactly when decay starts, and a normal losing streak can look identical in the early stages. That is why limits need to be set before you trade.
Tips for Beginners
- Count every cost in your backtest.
- Keep the latest data aside and test on it last.
- Go in stages: backtest first, then paper trade with virtual money, then trade live in a very small quantity.
- Write your stop rule before the first trade. For example, you might pause if your loss goes past the worst drawdown in your test. Choose your own number.
- Fix a review point in advance, such as every 30 trades. That number is only an example.
- Keep a trade journal with entry, exit, charges, and slippage.
- Prefer simple rules with few settings.
Strategy Decay vs a Normal Losing Streak
Point | Strategy decay | Normal losing streak |
What changed | The edge itself | Short-term luck |
Duration | Continues across many trades | Ends within the range you tested |
Drawdown | Deeper than your test | Inside the tested range |
Key numbers | Keep trending down | Stay close to the backtest |
What to do | Pause and retest | Follow the plan, keep size small |
Risk of a wrong call | Staying in too long | Quitting a working strategy early |
Common Mistakes to Avoid
- Trusting one backtest. It covers a single period. Test across rising, falling, and sideways markets.
- Leaving out costs. In the illustration above, ₹300 per trade turned a profit into a loss. Count every charge.
- Over-tuning settings. Odd values like a 43-day average often fit old noise. Choose simple values and check that nearby values also work.
- Judging too soon. Five losing trades say very little. Wait for a reasonable sample.
- Adding size after losses. Cut size instead until the strategy proves itself again.
- Copying a popular strategy. Crowded ideas fade fast, as the 920 Straddle shows.
- Never reviewing. A strategy left alone for a year may have changed without you noticing.
Who Should Watch for Decay?
Anyone trading on fixed rules should pay attention. That includes algo traders (whose rules are run by software), swing traders using chart setups, and F&O traders, where leverage makes small changes in edge matter more.
Long-term investors who hold good businesses for years face other risks, such as a weakening business or a high purchase price. Edge decay does not affect them in the same way. Traders with no fixed rules have nothing consistent to measure.
If you plan to automate your rules, start with how to set up your first algo trading system. To trade, you can open a demat account. Algo trading and backtested strategies carry market risk and do not guarantee returns.
Conclusion
Strategies decay because markets change, more traders pile into the same idea, and costs eat small profits. The 920 Straddle shows it with real numbers: the idea stayed the same while profit per trade fell. Count every charge in your tests, keep recent data aside, and compare live results with your backtest. Write down your stop point before the first trade. Trade small, pause when results fall outside what you tested, and retest before going back in. No strategy lasts forever, so plan for that from day one.
Frequently Asked Questions
Blog by Aliceblue News Desk
