SEBI published a study last year that should make every retail trader pause. In FY2024-25, 91% of individual traders in India's F&O segment lost money. Total losses crossed ₹1,05,603 crore — up 41% from the year before. Average loss per person: ₹1.1 lakh.
The strategies weren't the problem for most of them. The execution was.
Traders who know exactly what their setup says to do, and then don't do it — or do the opposite — aren't failing because the market is random. They're failing because the gap between knowing and doing is wider than they expected. That gap has a name: trading psychology. And it's where most retail accounts bleed out.
Why this topic matters
The SEBI data isn't new. The regulator flagged similar numbers for FY22-24, when 93% of individual F&O traders ended in the red. That's not a bad year. That's a consistent pattern across multiple years, across bull runs and corrections, across different market conditions.
If bad strategy was the cause, you'd expect some years to be better. They aren't, because the strategy usually isn't the problem. The mind is.
Most retail traders can tell you what they should have done after a bad trade. The setup was valid. The stop was clear. They just didn't take the loss, or they added to the position, or they re-entered because they were angry. This is what losing looks like in practice — not a broken strategy, but correct knowledge, incorrectly applied.
What emotional mistakes actually cost traders
FOMO is the most common entry error. The trade has already moved. Confirmation bias kicks in, the trader tells themselves there's more to come, and they enter at the worst possible point. The move usually reverses shortly after, and they're stuck holding a position with no edge.
Loss aversion does the most damage over time. Psychological research consistently shows that the pain of a loss registers roughly twice as intensely as the pleasure of an equivalent gain. In practice, this means traders hold losing positions far too long — because booking a loss feels final, while an open loss still has the possibility of reversing. The result is small profits and large losses, which is exactly backwards from what a working strategy produces.
Averaging down is loss aversion in action. When the trade goes against you and you add to it instead of cutting it, you're betting that being wrong twice is somehow better than being wrong once. Sometimes it works. When it doesn't, the loss is much larger than it needed to be.
Overconfidence tends to appear after a good run. After three or four winning trades, position sizes grow, stop losses widen, and the trader starts taking setups they'd normally skip. The market doesn't care about winning streaks, and the next loss — now on a larger position — wipes out the previous gains.
Revenge trading happens in the session after a bad loss. The trader is trying to recover ₹15,000 before the close, trading a setup that isn't there, in a size they shouldn't be trading. This is how one bad trade becomes three.
Decision fatigue is less discussed but real. By 1 PM after a full morning session, the quality of trading decisions degrades for most people. Discipline that held at 9:30 AM starts slipping. Stops that were clear earlier get moved.
How algo trading addresses each of these
The clearest thing an algorithm does is remove the gap between rule and action. The setup fires, the order goes in. There's no pause to reconsider, no looking at the P&L first, no waiting to see if maybe the trade comes back.
On FOMO: the algo takes the setup if it meets the criteria, and doesn't take it if it doesn't. It doesn't feel left out when a move happens without it. It just waits for the next valid signal.
On loss aversion: the stop loss executes automatically. There's no internal negotiation about whether to wait for one more candle. The exit happens at the price you decided before the trade, when you were thinking clearly.
On averaging down: rules-based position sizing means the position size is set before entry. If the strategy doesn't include averaging, it doesn't happen — because there's no human in the loop to override it.
On overconfidence: the algo doesn't have a winning streak. It runs the same logic on trade 47 as it ran on trade 1. No size creep, no loosened filters.
On revenge trading: there is no revenge. If the daily loss limit is ₹10,000, the system stops. The trader can be furious. The system doesn't care.
On decision fatigue: the algo runs the same logic at 2:45 PM as it ran at 9:20 AM.
One honest caveat here: emotional bias doesn't disappear when you switch to algo trading. It moves. The bias now enters during strategy design — if you build fear or greed into your rules, the algorithm executes fear or greed at scale, faster and more consistently than you could manually. Building a well-tested strategy is where the psychological discipline is still required.
Where AlgoSys can help
The AlgoSys platform gives traders two ways to reduce emotional interference, depending on how much control they want to keep.
Semi-automated mode is useful if you're not ready to hand over execution entirely. The signal fires, appears on your dashboard, and you approve it before it goes to the broker. This removes impulsive entries — you're not chasing a move in the heat of the moment — while keeping you in the decision loop. Stop losses and targets still execute automatically after entry.
Fully automated mode removes execution from the equation completely. TradingView alerts via webhook, AmiBroker AFL signals, or Python strategies route directly to your broker account through TradingView automation without any manual step.
The daily loss cap acts as a hard stop on the session. Set a limit, and the system halts when it's hit. This is the kill switch that most manual traders tell themselves they'll use, and then don't, because booking the loss and walking away is genuinely difficult. The cap removes that choice.
Kill switch gives you instant square-off of all open positions when you need to exit everything at once — not after deliberating, just immediately.
Practical checklist before automating
Before you hand execution to an algo, make sure the strategy has actually been tested:
- Backtest on at least 12 months of historical data, including periods of high volatility
- Define entry rules precisely — "when RSI crosses 30" is testable; "when it looks oversold" is not
- Set a fixed stop loss per trade and a daily loss limit before going live
- Paper trade for at least two weeks to check for gaps between backtest and live behaviour
- Start live with smaller position sizes than your backtest used
- Keep a log of every trade the system takes — not just the results, but whether the setup matched the rules
If the strategy depends on your judgment at entry or exit, it's not an algo yet. It's a semi-automated system, which is fine — but be honest about what you're actually running.
Common mistakes to avoid
Overriding the stop loss manually. If you're automating to remove emotional decisions and then turning off the automation when it's about to stop you out, you've built a system that only works when it agrees with your gut. That's not automation.
Mistaking a drawdown for a broken strategy. Every strategy has losing periods. Three losing trades in a row doesn't mean the strategy stopped working. Check whether the setup criteria are still being met correctly before making changes.
Changing the strategy during a losing run. This is the automated equivalent of revenge trading — tweaking the rules after losses to try to recover faster. Changes made under pressure are almost never improvements.
Going fully automated before understanding the strategy. If you don't know exactly why each rule exists, you won't know when to shut it off. Unexpected market conditions happen. A trader who understands the strategy can intervene; one who doesn't is flying blind.
Setting a daily loss limit and then disabling it. The limit only works if it's treated as fixed. If you override it once, you'll override it again.
FAQs
Does algo trading guarantee better returns than manual trading?
No. Algorithmic trading removes emotional interference from execution, which is one major source of underperformance. But the strategy still needs an edge to be profitable. An algo running a bad strategy loses money faster and more consistently than a manual trader would.
Can I use semi-automated trading if I'm not confident in full automation?
Yes. Semi-automated mode lets you approve each signal before it executes, so you stay in control of entries while the platform handles stop loss, target, and trailing SL automatically after you approve. Many traders start here before switching to full automation.
How do I know if my losses are due to emotional trading or a bad strategy?
Keep a trade journal. For each trade, note whether you followed the plan or deviated from it. If most losses happened when you deviated — held past the stop, entered late, sized up — that's an emotional execution problem. If you followed the plan and still lost, the strategy needs work.
What is a daily loss cap and how does it help?
A daily loss cap is a pre-set limit on how much the system can lose in a single session before it stops trading. It prevents a bad morning from turning into a catastrophic day. On AlgoSys, you set this limit in advance, and the system halts automatically when it's reached, regardless of what setups are showing.
Is trading psychology relevant for algo traders?
Yes, but it shows up differently. Manual traders fight psychology during execution. Algo traders fight it during design and monitoring — the temptation to override the system, change rules mid-drawdown, or disable safeguards. The discipline required is real either way.
How does AlgoSys handle risk management automatically?
AlgoSys supports pre-set stop loss, target, and trailing stop loss per trade, a daily loss cap, and a kill switch for immediate position exit. These execute automatically without manual intervention after the initial setup.
---
AlgoSys is a technology platform for trading automation. It does not provide investment advice, trading tips, portfolio management, or guaranteed returns. Trading involves market risk.