Day trading time window strategy: trade only between two clock times

A shape rather than a strategy. Most of my ideas only make sense in part of the day, and wrapping them in a window is the first thing I do.

1. The rule

# BUY — only between two times if HhmmSs >= 100000 and HhmmSs <= 143000: if Price > 0: # replace this line with your entry condition Buy()# SELL — force flat before the close regardless if HhmmSs >= 154500: Sell() elif ProfitPct >= 2.0 or ProfitPct <= -1.0 or HoldTime >= 1200: Sell()The forced flat is the part people forget. Decide whether you hold overnight deliberately rather than by omission.

2. Why a time window at all

Different parts of the session behave differently enough that one set of thresholds rarely fits all of them. The open is fast and expensive, the middle of the day is slow, and the last half hour has its own flows. Restricting entries to a window is the cheapest way to stop a strategy from being tested on one regime and traded in another.

3. When it fails

The window itself can be the overfit. If 10:00 to 14:30 tests better than 09:30 to 16:00, ask whether that is a real property of the market or a property of your sample. Try neighbouring windows — if 10:15 to 14:15 falls apart, the edge was in the boundary, not the idea.

4. What to change first

The forced close-out time. HhmmSs >= 155000 closes ten minutes before the bell. That is a deliberate margin: closing at 15:59 means competing with everyone else doing the same thing. Try 15:45 and see whether your fills improve enough to pay for the earlier exit.

5. What the round trip asks of it

The window is doing more work than it looks. A 2.0% target against a 1.0% stop needs a 33% win rate on paper and 58% with the measured 0.75% round trip (what day trading a US stock really costs). Twenty-five points is the toll for trading at all, and it is charged per trade. That is the strongest argument for a time window: every hour you exclude is a set of trades you do not pay for. A window that removes the flat middle of the session usually improves results not because those hours are unprofitable, but because they are *marginally* profitable and the round trip eats the margin. Which hours to keep is measurable rather than a matter of taste — see the opening 30 minutes is the easiest day trading window.