Time-Based Exit Strategy: Cut Stale Trades

Most traders have two exits, a stop and a target, and a time-based exit is the third one they forget. A stop answers “how wrong can I be?” A target answers “how right do I expect to be?” Neither answers “how long should this take?” Yet every trade has an implied answer to that question, and trades that overstay it rarely end well.

This post explains what a time-based exit is, why trades have a shelf life, when a time stop beats a price stop, how to find the right holding period from your own backtest data, and how to wire it into a strategy with visual blocks.

What Is a Time-Based Exit?

A time-based exit is a rule that closes a position after a set number of candles, regardless of profit or loss. It is sometimes called a time stop. If a trade opened on the four-hour chart with a limit of twelve candles, it closes two days later whether it is up, down or flat.

That sounds crude next to a carefully placed stop-loss. It is not crude. It is a different kind of rule. A price stop protects against being wrong about direction. A time stop protects against being wrong about timing, which is a separate and equally common failure.

Why a Trade Has a Shelf Life

Every entry signal is a claim about the near future. A breakout says “price will continue in this direction soon”. A mean reversion signal says “price will return to average soon”. The word “soon” is doing a lot of work in both sentences.

If the expected move does not arrive within the window the signal implied, the signal has expired. The market has had its chance to prove the thesis and has declined. What is left is a position with no reason behind it, held only because it has not yet hit the stop.

Mean reversion trades show this most clearly. Reversion to the mean is a fast process when it works. A stretched price snaps back within a handful of candles. When it does not snap back, the stretch has usually become the start of a new trend. Holding the trade at that point is not patience. It is hoping.

Breakouts follow the same logic with a different timeline. A real breakout has momentum in the first few candles. One that drifts sideways after the trigger has lost the participation that made it a breakout.

When Do Time Stops Beat Price Stops?

Price stops fail in one specific way. They wait for the market to prove you wrong, and the market can take its time doing that. A trade can sit a few percent underwater for days, never quite hitting the stop, tying up capital and attention the whole time.

A time stop closes that trade at a small loss or scratch instead of a full stop-out. Over many trades the saving is significant. The stalled trades that eventually hit the price stop are converted into smaller losses. The stalled trades that eventually recovered are lost, and that is the cost. The backtest tells you which group is larger.

Time stops also solve a problem price stops cannot see. Weekend and event risk. A position that has done nothing for eight candles heading into a Friday close is carrying overnight risk for no reason. A time rule takes it off before the thin liquidity arrives.

How Do You Find the Right Holding Period?

The right number comes from your own data, not from a rule of thumb. Two measurements do the job.

Take every trade in a backtest and record how many candles it took to reach its maximum favourable excursion, the best point the trade ever reached. Plot the distribution. For most strategies it is heavily front-loaded. A large share of trades peak within the first few candles and a long tail never really improves after that.

Then look at the winners separately from the losers. If 80% of winners reached their peak within six candles, holding beyond six is mostly exposing you to the losers. Set the time stop a little past that point. Not at it, because you do not want to cut the winners that are slightly late.

Run the same analysis on adverse excursion, the worst point the trade reached. Trades that are still underwater after the typical winner’s peak time are the ones the time stop is designed to remove.

Three Ways to Wire Time Into an Exit

Hard time stop. Close at N candles, no conditions. Simplest to build and to test. Start here.

Time plus no progress. Close at N candles only if the trade has not reached a minimum gain. A trade that is up 1R at the deadline gets to run. A trade that is flat or negative is closed. This keeps late winners while still removing dead trades.

Time-decayed target. The profit target shrinks as the trade ages. At candle one you want the full move. At candle ten you will accept half of it. This encodes the idea that a late move is a weaker move, and it exits earlier without a hard cut-off.

Test all three against the same entries. Keep the entry rules fixed and swap only the exit. The difference in expectancy is the value of the time rule, isolated from everything else.

How to Apply Time-Based Exits in Arrow Algo

Every version above is buildable with drag-and-drop blocks and no code.

The core is the Timer block. It starts counting when a signal fires and outputs true once a set number of candles has passed. Wire your entry signal into its start input and set the length to your chosen holding period. Connect its output to your exit. That is a hard time stop in three blocks. The Timer block guide covers its settings in detail.

For “time plus no progress”, add a comparison block checking unrealised profit against your minimum. Join it with the Timer output through a condition block set to AND. The exit fires only when the deadline has passed and the trade has not earned its keep.

For a decaying target, use a Counter block to track candles since entry and a subtract block to reduce the target by a fixed amount per candle. Compare price against the shrinking target.

Keep the price stop in place alongside all of these. A time stop is an addition to risk management, not a replacement for it. An ATR block feeding the stop distance keeps that side of the trade honest.

Then backtest with holding periods of 5, 10 and 20 candles. Arrow Algo pulls candles straight from Binance, Coinbase and HyperLiquid, so the stalled trades are all there in the data. Look at how many losses shrank, how many winners were cut, and what happened to the average holding time. Those three numbers tell you whether the rule earns its place.

What Are the Key Takeaways?

A time-based exit closes a trade after a fixed number of candles, whatever the profit or loss.
Every signal implies a timeline. When the move does not arrive on time, the signal has expired.
Time stops convert slow stop-outs into small losses and free capital for the next signal.
Find the holding period from the distribution of time-to-peak in your own backtest, not from a rule of thumb.
“Time plus no progress” keeps late winners while removing dead trades.
Keep the price stop. The time stop is a third exit, not a substitute for the first two.

Disclaimer: This content is for educational purposes only and does not constitute financial advice. Trading involves significant risk and you should only trade with capital you can afford to lose. Past performance is not indicative of future results. Always conduct your own research before making any trading decisions.

Ready to build your own automated trading strategies without writing a single line of code? Start for free at Arrow Algo and join thousands of traders who've made the switch to systematic trading.