Scaling, partials and stop placement

Every R:R table in Module 4 assumed the same management: half off at the first target, stop to breakeven, the rest held for the second target. That is one of several ways to manage a position and it is not automatically the best one. This lesson compares them, then turns to the stop itself: whether to set it by a fixed number of ticks or by chart structure, and what each choice costs.

Three ways to manage the same trade

Take a trade with entry, a stop 1R away, and two targets at 1R and 3R. Suppose that over many trades: 45% hit the stop before 1R, 30% reach 1R then return to breakeven, 25% reach 3R.

All-out at 1R. Win rate 55%, every winner is +1R. Expectancy = 0.55 × 1 − 0.45 × 1 = +0.10R

All-out at 3R, no partial. Only the 25% reach it; the other 75% are stopped at −1R (the 30% that reached 1R and returned are full losses here, since there is no breakeven stop without a partial). Expectancy = 0.25 × 3 − 0.75 × 1 = 0.00R

Half at 1R, breakeven stop, half at 3R. The 30% that return give +0.5R; the 25% winners give 0.5 + 1.5 = +2R. Expectancy = 0.45 × (−1) + 0.30 × 0.5 + 0.25 × 2 = −0.45 + 0.15 + 0.5 = +0.20R

The partial approach wins in this distribution because the 30% of trades that reach 1R and reverse are the swing factor. They are the most common intraday outcome and the partial converts them from losses into small wins. If your product produces very few of them (a strongly trending instrument on trend days), the all-out-at-3R approach can beat it; that is a Module 8 question for your own sample.

Scaling in (adding to a position that moves in your favour) is not covered here on purpose. It raises the average entry, concentrates risk at the end of the move, and is the mechanism behind most of the large single-trade losses in beginner logs. Add it, if at all, after the 100-trade sample.

Fixed-tick stops versus structure stops

Fixed-tick stops: the same distance every time. "8 points on ES, $0.30 on this stock." Predictable size, easy to preset in the bracket, and immune to the temptation to widen. The flaw is that the market does not know your number: an 8-point stop on a day with 20-point 5-minute bars is inside the noise.

Structure stops: placed beyond a chart feature: the opening range midpoint, the pullback low, the failed breakout high. The stop is where the setup is wrong, which is the correct logic. The flaw is that the distance varies, so size must be recalculated every trade, and a structure stop is easy to place "a little wider" to be safe, which is how a 1R stop becomes 1.6R.

The playbook uses structure stops, with a fixed-tick ceiling:

1. Find the structure stop for the setup (Module 4 gives it for each).
2. Compute the size: risk per trade ÷ stop in dollars per unit.
3. If the structure stop is more than 1.5× your product's normal stop, skip the trade. It is not a bad setup; it is a setup whose stop is inside a range too wide for your risk.
4. Add a noise buffer of about a third of the average 5-minute bar range beyond the structure, never more.

The normal stop is your own number, from your log: the median structure stop over your last 50 trades in that product. It changes with volatility, so recompute it monthly.

Where stops get hunted

Stops just beyond obvious levels (the round number, the exact opening range high, the exact prior-day low) are where every other retail stop sits, and price frequently trades a few ticks through the level before reversing. This is not conspiracy; it is that a cluster of stops is liquidity, and liquidity gets taken. Two adjustments:

• Place the stop beyond the level plus the noise buffer, not at the level.
• If the extra buffer takes the stop past the 1.5× ceiling, the trade is too tight for the setup and you skip it, rather than accept a stop at the level that you know is inside the usual overshoot.

  Structure stop with buffer

  OR high 100.40  ---------------------------
                              stop at level: hit by overshoot to 100.46
                            _/\
  overshoot zone           /   \  <- usual overshoot 1 to 6 ticks
                          /     \
  100.52 ----------------/-------\----- stop with buffer: survives
                                   \___ reversal

Breakeven stops: when and when not

Moving the stop to breakeven after the partial is standard in this playbook. Two exceptions:

• Not immediately. Moving to breakeven the moment the partial fills, while price is still within a few ticks of entry, converts a good trade into a scratch on ordinary noise. Move it when price is at least a third of the way from entry to the second target, or after the next 5-minute bar closes in your favour.
• Not on trend days for the runner. On a confirmed trend day, trail under each 5-minute higher low instead. A breakeven stop on a trend day is usually hit by the one deep pullback and then the trend continues without you.

Key idea: Manage the trade the same way every time, size from the structure stop with a fixed ceiling, and put the stop beyond the level plus a noise buffer. Consistency in management is what makes the R:R tables in your log mean something.

Try it: Take your last 30 trades (or 30 replay trades). Compute the expectancy under all-out at 1R, all-out at your second target, and the half-and-half approach, using your actual price paths. The best of the three for your product is your management rule for the next 100 trades.

Recap

• With a typical intraday distribution, half off at 1R and half at 3R with a breakeven stop beats all-out at either target.
• Do not scale in until the sample is done; it concentrates risk at the end of the move.
• Use structure stops, add a noise buffer of about a third of a 5-minute bar, and skip trades whose stop exceeds 1.5× your normal stop.
• Stops exactly at obvious levels sit in the overshoot zone; place them beyond it or skip.
• Move to breakeven only after price has travelled a third of the way to target two; trail instead on trend days.