General

What is Session Bias?

Reviewed by Karthikeyan Anbarasan (Founder, TradeLyser) · Updated 19 Oct 2026 · Editorial policy

DefinitionSession bias is a measurable tendency to trade better or worse in particular parts of the trading day, found by splitting your journal results by time window. An illustrative intraday trader who made +₹18,000 between 9:15 and 9:45 but lost ₹9,000 between 11:00 and 13:00, across the same 190 trades, has a clear bias towards the open.
Formula
Expectancy per window = Net P&L in the window ÷ Trades in the window (or average R per trade in the window)
Example
9:15–9:45: ₹18,000 ÷ 38 trades = ₹474 per trade; 11:00–13:00: −₹9,000 ÷ 52 trades = −₹173 per trade (illustrative)
+₹18k9:15–9:45+₹11k9:45–11:00−₹9k11:00–13:00−₹4k13:00–14:30+₹7k14:30–15:30
Figure 1. Illustrative net P&L by NSE session window: 9:15–9:45 +₹18k, 9:45–11:00 +₹11k, 11:00–13:00 −₹9k, 13:00–14:30 −₹4k, 14:30–15:30 +₹7k. The day nets +₹23k, with all of it made in the first and last windows.

How to measure it

The NSE day is not uniform. The first half hour after 9:15 usually carries the opening auction’s imbalances, overnight news and the widest ranges; late morning settles into trends or ranges; the middle of the day is often slower; and the last hour brings position squaring and MIS exits. Your own setups interact with each of these differently, so the same trader can have an edge at one time and give it back at another. Measuring it takes only the entry times already in your broker data.

  1. Use the entry time of each trade from your broker data, not times written from memory.
  2. Choose fixed windows that match how the market moves, such as the opening 30 minutes, late morning, midday, early afternoon and the last hour.
  3. For each window, total the trades, net P&L after charges and average R per trade.
  4. Check the trade count in each window; below about 30, treat the result as provisional.
  5. Compare windows on average R or rupees per trade, not on total P&L, since busy windows always have bigger totals.

Worked example

WindowTradesNet P&LPer trade
9:15–9:4538+₹18,000+₹474
9:45–11:0044+₹11,000+₹250
11:00–13:0052−₹9,000−₹173
13:00–14:3030−₹4,000−₹133
14:30–15:3026+₹7,000+₹269
All day190+₹23,000+₹121

Illustrative, after charges. The midday window holds 52 of 190 trades, about 27%, and loses money, which is enough trades to take seriously. The early afternoon has only 30 trades, so its result is weaker evidence. The last hour, with 26 trades, looks good but is still provisional. Without the two losing windows, the same trades would have netted +₹36,000, but that figure uses hindsight: the useful step is a forward test, such as no new entries between 11:00 and 14:30 for the next month.

Best window, per trade
+₹474
Worst window, per trade
−₹173
Share of trades at midday
27%
All-day average per trade
+₹121

Two meanings of session bias

Your performance bias

The pattern on this page: your own results differ by time of day. It is measured from your journal after the fact.
Answers: when do I trade well?

A market directional bias

A view, formed before or during the session, that the market is more likely to go up or down today, from cues such as GIFT Nifty, global markets and key levels.
Answers: which way am I leaning today?

What to do with it, and when it misleads

Treat a session bias as a hypothesis to test, not a verdict. Test sitting out the weakest window for a month, or trading it at half size, and compare the next month with this one. Split each window by setup tag: an opening range breakout only exists near the open, so a strong 9:15 window may be a strong setup rather than a strong time. Check whether losses in a window follow earlier losses in the day, since midday losses are often the second or third attempt after a bad morning, which is a revenge or overtrading problem rather than a clock problem. Review session results in the monthly review, alongside expectancy per strategy, rather than after one bad day.

  • One very large day can make a window look good or bad on its own; check the result without the best and worst day.
  • Windows with fewer than about 30 trades carry wide margins, the same as any small sample.
  • Market character changes: a window that was strong in a high-volatility period may be quiet in a calm one.
  • Time can be a stand-in for something else, such as fatigue, trading after a loss or trading on expiry day.
For traders in India
NSE’s pre-open session runs from 9:00 to 9:15 and normal trading from 9:15 to 15:30, with MIS positions squared off by the broker before the close. Nifty weekly options expire every Tuesday, and expiry afternoons often trade differently from other days, so read Tuesdays as a separate group. TradeLyser’s analytics can break results down by session, and cross analysis compares sessions, strategies and symbols side by side.

Reference guide

ContextValueReading
A weak window with 30 or more trades, tested by sitting out for a monthTested changeThe next month shows whether skipping the window actually helped
A window judged on six tradesNoiseSplit results by time only once each window has enough trades to compare
Losses in a window that hold only one setupHidden causeCheck whether the time or the setup is the problem before banning the window

Questions

What is session bias in trading?

A consistent difference in your results between parts of the trading day, such as making money near the open and losing it at midday. It is found by splitting journal results by time window.

How do I find my best time of day to trade?

Group your trades by entry time into fixed windows, then compare average R or rupees per trade in each, after charges. Trust only windows with around 30 trades or more.

Is the first 15 or 30 minutes after the open the best time to trade?

It is often the most active part of the day on NSE, but whether it is best for you depends on your setups. Your journal, split by window, is the only reliable answer.

Should I stop trading during my worst session?

Test it rather than assume it: sit out or halve size in that window for a month and compare. Also check whether the losses come from one setup or from trading after earlier losses.

What are NSE trading hours?

Pre-open runs from 9:00 to 9:15 and normal trading from 9:15 to 15:30 on trading days. Check exchange notices for special sessions and holidays.

Related terms