What it is
A trend day is a session in which the market opens near one extreme, moves steadily in one direction with shallow pullbacks, and closes near the other extreme. They are a minority of sessions, perhaps 15 to 25 percent depending on the market and the year, but they produce a large share of the directional profit available to day traders. This playbook is about recognising a trend day in the first hour, holding a position through the pullbacks that shake most traders out, and not fading anything.
It is written for index futures but applies to any liquid futures contract, forex pair or crypto perpetual.
The logic
Trend days occur when a large participant (or many aligned participants) has a lot of business to do in one direction and does it across the whole session. The pullbacks are small because the responsible buyer or seller is patient enough to work into every dip, and the traders who fade the move provide liquidity to them until they are forced to cover. The close near the extreme reflects late-day capitulation by the fade traders and market-on-close flows in the same direction.
The other side on a trend day is every mean-reversion strategy, every range trader and every "it's overextended" short. On a trend day they are all wrong, and their covering is the trend's fuel. The playbook's edge is patience: holding when the natural impulse is to take profit.
Setup rules
- Market: es, nq, cl, treasury futures; major forex pairs during the london-session and new-york-session overlap; liquid crypto perpetuals.
- Timeframe: 5-minute for structure, 15 or 30-minute for the trend definition.
- Early tells (need at least three by 10:30 ET): open outside the prior day's range or a gap that does not fill in the first 30 minutes; the first 30-minute bar is a wide-range bar closing near its extreme; price stays on one side of vwap and VWAP slopes; the initial-balance-extension occurs early and holds; relative-volume above 1.3x; pullbacks fail to close beyond the 20-period EMA on the 5-minute chart; cumulative delta (if you have it) matches direction.
- Context: a scheduled macro catalyst overnight or at the open, or a large gap on index-level news, raises the prior probability.
- Disqualifiers: the day is inside the prior day's range after 60 minutes; VWAP is flat; price crosses VWAP more than twice in the first hour.
Entry, stop, target
Enter on the first pullback to the 5-minute 20-period EMA or to VWAP that holds (a bar closing back in the trend direction). Stop beyond the pullback's extreme, or beyond VWAP if that is wider. There is no fixed target; the plan is to hold to the last 15 minutes and exit into the close, trailing the stop under each successive 15-minute swing low. Add once at the second pullback if the first has moved 1R in your favour, not more.
| Item | Level | Notes |
|---|---|---|
| Open | 5,010.00 | Gapped above prior day high |
| 30-minute bar | 5,010 to 5,032, closed 5,030 | Wide range, closed near high |
| First pullback to EMA | 5,022.00 | Held, closed back up |
| Entry | 5,025.00 | Close of the holding bar |
| Stop | 5,017.00 | Below pullback, risk 8 points |
| Trailed exit at 15:45 | 5,061.00 | Reward 36 points, 4.5R |
The 4.5R is what a good trend day pays; the losses on days that looked like trend days and were not are typically 1R each, and there are more of those. The expectancy comes from the ratio.
Position sizing and risk
One position sized at 0.5 to 1 percent of equity from the pullback stop, computed at /tools/position-size, with micro-futures used to hit the number exactly. The add at the second pullback is the only scaling allowed, and the combined position must still respect the daily loss limit in /learn/risk-management. Because trend days are rare, the temptation to over-size "the one that's finally working" is strong; the size is set by the stop, not by the excitement.
What breaks it
- Failed trend days. A morning that has every tell and reverses at 11:00 ET. These are the cost of the playbook and are unavoidable; the stop takes a 1R loss.
- Midday chop. Even genuine trend days often go sideways for two hours around lunch; traders exit out of boredom and miss the afternoon leg. The playbook's rule is to hold until the trailed stop is hit or the close arrives, not until it gets dull.
- Costs. Minor; one or two round trips per day in liquid futures.
- News reversals. A scheduled release at 14:00 ET can end a trend day instantly; know the calendar and tighten the stop into it.
- Edge decay. Trend-day recognition is not a secret, but holding through pullbacks is behaviourally hard enough that it remains underexploited. The decay risk is lower than for pattern-based setups.
- Drawdowns. Weeks without a single trend day are normal; the playbook needs another source of trades (see range-day-playbook) or the patience to sit out.
How to test it
Classify at least 300 sessions by the close's position in the day's range (top or bottom 15 percent equals trend day) and record which of the early tells were present by 10:30 ET on each. That gives you the predictive value of each tell, which is the real research. Then replay the sessions that met your threshold with your entry and trailing rules and record R outcomes. Expect a win rate near 40 percent with an average winner several times the average loser; if your test shows a 70 percent win rate, the classification is leaking the outcome. Simulate for 30 sessions before trading live.
Variations
- Trend-day fade avoidance as a standalone rule: on any day with the tells, disable all mean-reversion setups. That alone improves many traders' results.
- Overnight-trend continuation where the Globex session already trended and the tells are checked at the open.
- Power-hour continuation: enter only the afternoon pullback in a confirmed trend day; fewer trades, tighter stop.
Further reading
trend, vwap, relative-volume, opening-range, market-on-close, es, nq, micro-futures, whipsaw, expectancy.
Related playbooks: range-day-playbook, initial-balance-extension, vwap-reclaim-reject, afternoon-reversal