Time of Day Performance
A strong Time of Day Performance shows at least 60% of net profits generated in 2-3 peak hours, with win rate and expectancy measurably higher in those windows than off-peak hours.
No credit card required
Benchmark Ranges
| Level | Range | What It Means |
|---|---|---|
| Excellent | One or two 1-hour windows account for 50%+ of net profit with win rate above 55% in those windows | Clear, exploitable time-based edge — concentrate trading in peak hours |
| Good | 2-3 hour windows account for 40-50% of net profit with win rate 50-55% | Moderate time edge present — worth filtering your sessions around these windows |
| Neutral | Profit roughly evenly distributed across 4+ hour windows, no window above 30% | No measurable time-based edge — focus on other performance variables |
| Poor | Net losses concentrated in specific hours you are actively trading | You are trading your worst hours — immediate session restriction recommended |
How to Track
Log the exact entry time (to the minute) on every trade in your journal
Group trades into 1-hour buckets aligned to market sessions (London open 08:00 GMT, NY open 13:00 GMT, etc.)
Calculate net pips, win rate, and average R:R for each hourly bucket across at least 50 trades
Identify the top 3 hours by net profit and compare their win rates to your overall win rate
Review quarterly — time-based edges shift with market regimes and seasonal volatility
How to Improve
Stop trading your worst 2 hours — if 14:00-16:00 GMT shows consistent losses across 30+ trades, block that window entirely
Increase position size during your top-performing hour by 25-50% to extract more from your proven edge
Align pairs to sessions — trade EUR/USD and GBP/USD during London hours, USD/JPY during Tokyo-London overlap
Track news events by hour and remove any trades taken within 15 minutes of high-impact releases from your baseline
Set a hard stop-trading rule after two consecutive losses in any session to prevent revenge trading distorting your hourly data
Time of Day Performance measures how your trading results — net pips, win rate, and expectancy — vary across different hours of the trading day. Rather than treating your performance as a single aggregate, this consistency metric segments your trade history into hourly windows to reveal when your edge is statistically strongest and when you are most likely to give back profits.
Formula & Calculation
Hourly Net Pips = Sum of pip P&L for all trades entered in hour H
Hourly Win Rate = (Winning trades in hour H / Total trades in hour H) × 100
Hourly Expectancy = (Win Rate × Avg Win Pips) − (Loss Rate × Avg Loss Pips)
Where:
- Hour H = a 1-hour UTC time bucket (e.g., 08:00-08:59 GMT)
- Win Rate = percentage of trades closed at a profit in that hour
- Avg Win Pips = mean pip gain on winning trades in that hour
- Avg Loss Pips = mean pip loss on losing trades in that hour (expressed as positive)
The calculation is run independently for each hour bucket, then compared across the full trading day to identify outliers — both positive and negative.
Benchmarks
| Level | Range | What It Means |
|---|---|---|
| Excellent | 1-2 hour windows account for 50%+ of net profit, win rate above 55% in those hours | Clear, exploitable time-based edge — concentrate trading in peak hours |
| Good | 2-3 hour windows account for 40-50% of net profit, win rate 50-55% | Moderate time edge present — worth filtering your sessions around these windows |
| Neutral | Profit roughly evenly distributed across 4+ hours, no window above 30% | No measurable time-based edge — focus on other performance variables |
| Poor | Net losses concentrated in hours you are actively trading | You are trading your worst hours — immediate session restriction recommended |
Practical Example
A trader runs 90 trades over 3 months on EUR/USD, logging entry times in UTC. Breaking trades into hourly buckets reveals:
- 08:00-09:00 GMT (London open): 18 trades, +312 pips net, 61% win rate, 1.8R average
- 13:00-14:00 GMT (NY open): 15 trades, +198 pips net, 60% win rate, 1.6R average
- 10:00-12:00 GMT (London midday): 22 trades, −47 pips net, 45% win rate, 0.9R average
- 16:00-18:00 GMT (late NY): 19 trades, −89 pips net, 42% win rate, 0.8R average
- Remaining hours: 16 trades, +31 pips net
Total net pips: +405. The London open and NY open together account for +510 pips — 126% of net profit — meaning all other hours combined are a net drag of −105 pips. This is an Excellent result: two hours produce the entire edge, while 68 trades outside those windows actively erode it. The clear action is to restrict live trading to 08:00-09:00 and 13:00-14:00 GMT.
How to Track Time of Day Performance
- Log exact entry time on every trade — to the minute, in UTC. Do not use local time, which shifts with DST.
- Group trades into 1-hour UTC buckets — align them to known session boundaries (London 08:00, NY 13:00, Tokyo 00:00 GMT).
- Calculate net pips, win rate, and expectancy for each bucket — require at least 20 trades per bucket before drawing conclusions.
- Rank hours by net pips and flag outliers — identify your top 2-3 hours and your bottom 2-3 hours.
- Review quarterly — time-based edges shift with market regimes; recalibrate session filters every 60-90 days.
How to Improve Time of Day Performance
- Cut your worst 2 hours from live trading — if 10:00-12:00 GMT shows a net loss across 25+ trades, block that window for 30 days and measure the impact on overall P&L.
- Increase position size during your top hour by 25% — if 08:00-09:00 GMT has an expectancy of +8 pips per trade, adding 0.25 lots to your standard size extracts significantly more from your proven edge without adding new risk.
- Align currency pairs to sessions — trade EUR/USD and GBP/USD during London, USD/JPY during Tokyo-London overlap, USD/CAD during NY. Pair-session mismatch is a common source of poor hourly performance.
- Remove high-impact news trades from your hourly baseline — a single 150-pip news spike can make a mediocre hour look excellent. Tag and separate news trades so your structural edge is visible.
- Apply a two-loss session stop — once you hit two consecutive losses in any session, stop trading for that hour. Revenge trading after losses inflates your worst-hour data and compounds the damage.
Common Mistakes
- Drawing conclusions from too few trades — an hourly bucket with 8 trades and a 75% win rate means nothing. Each bucket needs 20+ trades before the numbers are actionable.
- Treating all pairs identically — GBP/JPY volatility peaks at the London open differ from EUR/USD peaks. Run Time of Day Performance analysis separately for each pair you trade regularly.
- Including news outliers in hourly averages — one 200-pip flash move can make a structurally weak hour appear profitable. Always tag news trades and analyze them separately.
- Ignoring DST transitions — London, New York, and Tokyo shift clocks independently. If you log local time, a single DST event splits one trading window across two different UTC hour buckets, corrupting months of data.
- Optimizing for win rate alone — a 40% win rate at 3:1 R:R produces an expectancy of +0.8R per trade, which outperforms a 60% win rate at 0.8:1 R:R (expectancy +0.08R). Always evaluate hourly win rate alongside average risk-reward ratio.
How PipJournal Calculates Time of Day Performance
PipJournal automatically segments your trade history by entry hour in UTC, displaying an hourly heatmap on the analytics dashboard that shows net pips, win rate, and trade count for each window. The chart highlights your top and bottom performing hours using color intensity, so patterns surface immediately without manual calculation. You can filter by currency pair, date range, or session to isolate your EUR/USD London-open performance versus your overall dataset. All trade times are stored in UTC at the point of logging, eliminating DST distortion from your historical data. The session P&L breakdown view complements the hourly heatmap by grouping results into the four major forex sessions for traders who prefer a higher-level view.
Common Mistakes
Using too small a sample — hourly buckets need at least 20 trades each before conclusions are valid
Confusing pair-specific timing with universal timing — GBP/JPY volatility peaks differ from EUR/USD peaks
Including news-spike outliers in hourly averages — one 200-pip news trade can skew an entire hour bucket
Ignoring DST shifts — London open moves by 1 hour seasonally, which misaligns your historical buckets if not corrected
Optimizing only on win rate and ignoring average pip size — an hour with 40% win rate and 3:1 R:R can outperform a 60% win rate at 1:1
Frequently Asked Questions
What is Time of Day Performance in forex trading?
Time of Day Performance measures your profitability, win rate, and expectancy broken down by the hour of day you enter trades. It reveals whether you have a statistically meaningful edge during specific market hours — such as the London open, the NY session, or the Tokyo-London overlap — versus hours where you consistently underperform or lose.
Which trading hours are typically most profitable for forex?
The London open (08:00-10:00 GMT) and the New York open (13:00-15:00 GMT) are historically the highest-volatility windows for major pairs like EUR/USD and GBP/USD. However, the best hours vary significantly by pair, strategy, and individual trader. Your personal Time of Day Performance data is more useful than general market statistics.
How many trades do I need before my hourly data is reliable?
Each 1-hour bucket needs at least 20 trades before drawing conclusions, and 30-50 trades for statistical confidence. If you take 3-4 trades per day, expect to need 2-3 months of consistent data before hourly breakdowns are meaningful. Pooling data across similar market regimes (trending vs. ranging) improves reliability faster than just waiting for more trades.
Should I stop trading entirely outside my best hours?
Not necessarily — but it is worth paper-testing a session filter first. Block your worst 1-2 hours from live trading for 30 days and compare your net result. Many traders find that cutting their 2 worst hours improves overall performance even though it reduces trade count, because those hours were generating net losses that offset profits from better sessions.
How do I account for DST changes in my hourly data?
Convert all trade times to UTC and keep them in UTC throughout your analysis. Clocks in London, New York, and Tokyo shift independently during daylight saving transitions in March and November. If you log local time, a single DST event can split what is effectively the same trading window across two different hour buckets in your data.
Does Time of Day Performance matter for swing traders?
It matters less for swing traders holding positions for days or weeks, since entry timing has less impact on outcome than it does for intraday traders. For swing traders, session-level analysis (Asian, London, New York) is more relevant than hourly breakdowns. The metric is most actionable for day traders and scalpers who open and close within the same session.
Can Time of Day Performance change over time?
Yes — intraday volatility patterns shift with market regimes, central bank calendar cycles, and changes in institutional participation. A window that was consistently profitable during a trending EUR/USD regime may flatten out during a ranging period. Review your hourly breakdown quarterly and recalibrate session filters accordingly.
Track Your Metrics With PipJournal
Automatically calculate and track all your trading metrics in one place. See what's working and what's not.
Start Free TrialNo credit card required