Managing a setup: stops, targets and the time limit

A setup tells you when to get in. Management tells you where you are wrong, where you get paid and how long you are willing to wait. Every strategy in this series uses the same management rules, so they are explained once here. They are not magic numbers. They are one consistent, testable way to keep losses small, let some winners run and stop a trade from drifting on forever.

Two words first: ATR and R

ATR (average true range) is the typical size of a candle over recent candles, including any gap from the previous close. If the ATR on a 1-hour chart is 200 dollars, a normal hourly candle covers about 200 dollars from high to low. Measuring distances in ATR lets the same rules work on a quiet coin and a wild one.

R is the amount you risk on the trade: the distance between your entry and your stop. If you buy at 100 and your stop is at 98, then 1 R is 2. A target 3 above the entry is 1.5 R away. Thinking in R makes trades of different sizes comparable.

Entering at the close

Every setup is confirmed only when a candle closes. Until then the candle can still change shape: a long lower wick can turn into a plain red candle in the last minute. So the entry is at the close of the candle that completes the conditions, usually with a market order right after it closes.

It is tempting to place a limit order a little lower to save a few ticks. The problem is that the best trades often do not come back. A limit order that never fills means you skip the winners and still catch the ones that drift down to you first. If you test the rules with entries at the close, trade them that way too.

Where the stop goes, and why it has a buffer

The stop sits just past the structure the setup is built on: below the wick that swept a low, below a pullback, below a gap. If price goes there, the idea behind the trade is wrong. On top of that the stop gets a buffer of 0.2 ATR. Obvious lows attract other traders' stops, and price often pokes a little past them before turning. The buffer keeps you out of that small poke without moving the stop so far that it stops meaning anything.

Too tight, too wide

The stop must be between 0.25 and 4 ATR from the entry. A stop closer than a quarter of a normal candle is inside the noise. Ordinary wiggles will hit it even when the idea was right. A stop more than 4 ATR away is a different trade: it needs a much longer holding time and a much smaller position to keep the same risk. When a setup produces a stop outside this band, the trade is skipped rather than adjusted.

The fee cap, and why it bites on 15-minute charts

Assume fees of 0.1% to get in and 0.1% to get out. That cost is a fixed share of the price, but R depends on how far the stop is. On a tight stop, fees become a large slice of what you risk. The rule: fees may cost at most 0.3 R.

An example. Buy at 100 with a stop at 99.5, so 1 R is 0.5. Round-trip fees are about 0.2, which is 0.4 R. That fails the cap. On 15-minute candles stops are often this tight, so many setups get filtered out there. On very short timeframes with tight stops, much of what you make goes to the exchange.

Target 1 must pay at least 1.5 R

Each setup names its own target 1: a range middle, a swing high, a band. It has to be at least 1.5 R from the entry. If it is closer, the trade is skipped. A trade that risks 1 to make 0.8 needs to win far more often than most setups do, and that is before fees.

Half off at target 1, stop to entry

When price reaches target 1, half the position is closed and the stop on the rest moves to the entry price. Banking half locks in part of the move. Moving the stop to entry means the remaining half can no longer turn the trade into a full loss. The cost is that some trades get stopped at entry and then go on without you. That is accepted in exchange for a smoother result.

What target 2 is

Target 2 is the next level above target 1 that is at least 0.5 R further away. The levels counted are swing highs (a candle whose high is above the 3 candles before and after it), the volume profile's POC, VAH and high-volume nodes (prices where a lot of volume traded over the last 96 candles), the previous day's high, and the 24-hour and 7-day highs. The rest of the position comes off there. If no such level exists, everything comes off at target 1.

The 32-candle time limit

Whatever is still open is closed after 32 candles: 8 hours on 15-minute candles, 32 hours on 1-hour candles, about 5 days on 4-hour candles. A setup is a bet that something happens soon. If neither the stop nor a target is reached in that time, the reason for the trade has faded, and the money is better freed for the next one.

Test it with fees before trusting it

A rule set like this can and should be backtested on the coin's own recent history, with fees included and entries at the close. None of these rules is known to make money. A test will not prove a strategy works in the future, but it will often show quickly that one does not.

What to check

1. Did the candle actually close? Do not enter on a candle that only looks right halfway through. Wait for the close, then act on it.
2. Is the stop inside the 0.25 to 4 ATR band? Measure it in ATR before you size the position. If it falls outside, skip the trade instead of nudging the stop to fit.
3. What do fees cost in R? Work out round-trip fees divided by the entry-to-stop distance. If it is above 0.3 R, the trade mostly pays the exchange, especially on 15-minute charts.
4. Is target 1 at least 1.5 R away, with nothing heavy in between? A clear resistance level halfway to the target can stop the move early. Check the path, not only the destination.
5. Is the position sized from R? Decide how much money you are willing to lose, then divide by the stop distance. The stop sets the size, not the other way round.
6. Do you know your target 2 and your time limit before entering? Write down the candle on which the trade will be closed. Decisions made in advance are easier to follow than ones made mid-trade.
7. Did your backtest include fees and close entries? A test with free trading or perfect limit fills will look better than reality. Test the version you will actually trade.