Sleep, Fatigue and Night Trading: What Tired Brains Do and Rules for Thai Time

Thai traders have a disadvantage that rarely gets discussed. The busiest hours of the market and the big US news releases fall between the evening and 2 a.m. Thai time, exactly when the body should be resting. This article pulls together what sleep research knows with reasonable confidence about how fatigue affects decisions, then turns it into simple rules that fit a real schedule. You don’t have to give up trading at night, but you should know what kind of brain you are trading with.

This is not medical advice. If you have ongoing trouble sleeping, see the section near the end.

How many hours does the market overlap your sleep

Take each city’s business hours as about 08:00–17:00 local time (Tokyo 09:00–18:00), convert them to Thai time as in Forex market hours in Thai time, and set them next to an example sleep window of 23:00–07:00 (8 hours).

Time of year	London	New York	New York inside the sleep window
Winter (about November to mid-March)	15:00–24:00	20:00–05:00	23:00–05:00 = 6 hours
US and UK summer time	14:00–23:00	19:00–04:00	23:00–04:00 = 5 hours

US data such as NFP and CPI come out at 08:30 New York time, which is 20:30 Thai time in winter or 19:30 in summer time. The FOMC (the US Federal Reserve’s policy committee) decision comes at 14:00 New York time, which is 02:00 or 01:00 the following night in Thai time, with a press conference half an hour later. How to read the release schedule is covered in How to read an economic calendar.

In 2026 the UK leaves summer time on 25 October but the US does not until 1 November, so for that week the two sessions don’t shift together. If your stop-trading rule is tied to market hours, check that week carefully.

Three things sleep loss does to the brain

Most sleep research is done in labs, not on traders. But these three findings have been seen often enough to be worth using.

1. Attention drops out for moments, without you noticing. A tired person isn’t just evenly slower. They have short lapses where the brain simply drops out. In reaction-time tests these lapses rise with hours awake. Williamson and Feyer (2000, Occupational and Environmental Medicine) found that after 17–19 hours awake, performance on some tests was as bad as or worse than at a blood alcohol level of 0.05%, which is Thailand’s drink-driving limit for ordinary drivers. That matches an earlier result by Dawson and Reid (1997, Nature).

On a chart, a lapse is typing the lot size one digit off, placing the Stop loss on the wrong side, clicking buy instead of sell, or not noticing that news is due in five minutes.

2. Gains loom larger than losses. Venkatraman and colleagues (2007, Sleep) gave people a risky-choice game after one night without sleep. They shifted from protecting against losses toward seeking gains, with the brain responding more to gains and less to losses. This was a lab game, and it doesn’t mean everyone trades more riskily by the same amount. But the direction matches what many traders report: late at night they add size or decide to “hold a little longer”.

3. Emotional reactions get stronger. Yoo and colleagues (2007, Current Biology) found that after sleep deprivation, the brain region that reacts to negative stimuli fired more strongly and was less connected to the frontal areas that help us pause and think. A loss at 1 a.m. can therefore feel worse than the same loss at 3 p.m., and push you toward revenge trading, which is covered in Daily loss limits and revenge trading.

The most dangerous part is that tired people tend to rate themselves as fine. Van Dongen and colleagues (2003, Sleep) had people sleep 6 hours a night for two weeks. Their attention scores kept getting worse until they were close to people who had missed one to two full nights of sleep, yet their own sleepiness ratings rose only slightly.

Feeling “still fine” can’t measure readiness. Let the clock and last night’s hours of sleep decide instead.

One more thing worth knowing: for people who normally sleep at night, alertness from the body clock is usually lowest from around 2 a.m. to 6 a.m. An FOMC decision at 01:00–02:00 therefore meets both a long time awake and the low point of the body clock.

How far does “decision fatigue” go

Many articles cite Danziger and colleagues (2011), which reported that judges granted parole less often as their food break approached, as proof that decisions drain “willpower”. That study has been heavily challenged, because the order of cases that day wasn’t random, and cases likely to be refused may have been scheduled at the end of a session anyway. The broader idea that willpower is a battery that runs down (ego depletion) also failed to replicate in a large study run across many labs at once.

The careful summary: there’s no strong evidence that making many decisions by itself makes later decisions worse, but fatigue from being awake a long time is far better supported. The practical advice is the same either way. Make the big decisions while your brain is fresh, then write them down as rules so the late-night you only has to follow them.

Pick a session that fits your life

You don’t need to trade the busiest hours. A good session is one where you’re fully awake and can follow your plan every day.

Your schedule	Session worth trying	Keep in mind
Day job, bed at 22:00–23:00	19:00–22:00, the start of the London and New York overlap	Stop before bed by rule, and don’t hold new trades through late news
Early riser	Morning to afternoon, the Asian session	Many pairs move less; watch the spread around the broker’s daily rollover
No time to watch screens	Daily or H4 charts, checked once a day	Place orders with a Stop loss and go to sleep
Running an EA	Let it run on a server	Set alerts only for things that need urgent fixing; see Running an EA on a VPS

The “no new trades after X o’clock” rule

This rule works because you decide only once, while your brain is fresh.

1. Set X to one hour before bedtime. Bed at 23:00 means X = 22:00.
2. After X you may do only three things: close a trade, reduce a trade, or leave a trade that already has a Stop loss. No new trades and no moving a Stop loss further away.
3. Any trade held overnight must have a real Stop loss in the system, not one in your head.
4. On nights with major news after bedtime, decide before X whether to close, cut size, or hold through at a size whose outcome you can accept. Fast news can fill a Stop loss at a worse price than you set; the mechanics are in News spikes and slippage.
5. Add this rule to your pre-trade checklist. “How many hours did I sleep last night?” belongs there too.

If you decide to get up for an FOMC decision now and then, treat it as getting up to watch, not to rethink. For a while after waking the brain is still groggy (sleep inertia), and on the next day, short on sleep, cut your size or take the day off.

Orders and alerts instead of waiting up

Many people stay up to wait for price to reach a level they planned, when the platform could wait for them.

• Pending order: an order that waits to enter at a set price. Attach the Stop loss and Take profit at the same time and set an expiry so the order doesn’t sit there and fill during the news. Order types are explained in Order types explained.
• Price alerts work well while you’re awake, but an alert that wakes you at 2 a.m. is the same as not sleeping. Silence them during your sleep hours, or keep only the ones that really need action, such as margin getting close to a stop-out.
• If a plan needs someone to watch price all night, ask whether that plan suits a person who has to get up for work in the morning.

Caffeine and screen light

The amount of caffeine in the body halves in about 5 hours on average (its half-life), though people differ a lot. Take an assumed dose of 200 milligrams (the amount per cup varies widely by shop and size) and a bedtime of 23:00.

Caffeine left = 200 × 0.5^(hours elapsed ÷ 5)

Taken at	Hours until 23:00	Caffeine left
14:00	9 hours	200 × 0.5^1.8 ≈ 57 mg
17:00	6 hours	200 × 0.5^1.2 ≈ 87 mg
19:00	4 hours	200 × 0.5^0.8 ≈ 115 mg
22:00	1 hour	200 × 0.5^0.2 ≈ 174 mg

Check: from 19:00 to midnight is exactly 5 hours, so half should remain, 100 mg, and by 02:00 about 76 mg is left. The coffee that kept you up for the evening news is still in your body while you try to fall asleep. Drake and colleagues (2013, Journal of Clinical Sleep Medicine) found that a large dose of caffeine taken 6 hours before bed could still disturb sleep.

Screen light at night delays the body’s release of melatonin. Chang and colleagues (2015, PNAS) found that reading on a tablet before bed meant falling asleep later and feeling less alert the next morning than reading a printed book. The effect was measurable but not large. For traders, the bigger problem than the light is what is on the screen: a profit and loss figure that keeps moving wakes you up more than the light does. Checking your account often also shows you red numbers more often than you need to see them; more in Checking frequency and loss aversion.

In practice: close the charts 30–60 minutes before bed, keep the phone outside the bedroom if you can, and don’t check your account the moment you wake in the night.

Keep a sleep log next to your trading journal

Most adults need at least 7 hours of sleep a night on a regular basis, according to the joint recommendation of the American Academy of Sleep Medicine and the Sleep Research Society (2015). The way to see how much sleep affects your own trading is to add a few columns to your trading journal.

Date	Bed	Woke	Hours slept	Last coffee	Traded after X?	Off-plan trades
Example	01:30	06:30	5.0	20:00	Yes	2 of 3

An off-plan trade is one entered without all the conditions, sized above your rule, or with a moved Stop loss. It isn’t the same as a losing trade. A trade that followed the plan and lost counts as correct.

A made-up example: after four weeks, 20 trading days, split by hours slept the night before.

• Nights under 6 hours: 6 days, 9 trades, 4 off-plan = 4 ÷ 9 ≈ 44%
• Nights of 6 hours or more: 14 days, 21 trades, 3 off-plan = 3 ÷ 21 ≈ 14%

These numbers are invented to show how to read the log. Nine trades are far too few to conclude anything firmly, and short-sleep days may also be days stressed by something else. Keep logging until you have enough data; how many is enough is covered in How many trades before you can conclude anything. If the same pattern keeps showing up, you get a personal rule backed by your own data, for example: under 6 hours of sleep means half your normal size that day.

If you keep sleeping badly

An occasional bad night is normal. But if you struggle to fall asleep, wake in the night or wake unrefreshed for several weeks in a row, or snore so heavily that someone notices you stop breathing at times, see a doctor. Insomnia has effective treatments: cognitive behavioural therapy for insomnia (CBT-I) is the approach medical guidelines recommend first. Don’t use alcohol as a sleeping aid. It may help you fall asleep faster, but it makes the second half of the night’s sleep worse.

If you can’t sleep because you keep thinking about money lost trading, or the stress is affecting your life, help is available. In Thailand, the Department of Mental Health hotline is 1323. Signs that trading has stopped being healthy are covered in Burnout and gambling warning signs.

If trading regularly costs you sleep, fix the trading schedule first, not the coffee.

Common mistakes

• Trading the busiest hours because they must be the best, even though they fall in your sleep time. The best session is the one where you can follow the plan.
• Using your feelings to judge whether you’re fine. Research shows tired people overrate themselves. Use the clock and your hours of sleep instead.
• Waiting up for a price instead of placing an order. If the entry is clear enough to wait for, it is clear enough for a pending order with a Stop loss.
• Getting up for 2 a.m. news and opening a trade on impulse, when it wasn’t in the plan you made before bed.
• Forgetting that market and news times shift twice a year. Your “after X o’clock” rule needs a fresh look whenever summer time changes.
• Drinking coffee to stay up for the evening news, then wondering why sleep won’t come.
• Checking your account in bed. Moving numbers wake you up more than you think.