Three candidate families, three different questions
Choose a family before tuning thresholds. The following are illustrative research definitions, not a ranking or an assertion that each has a profitable edge.
| Family | Candidate condition to define | Missing entry decision |
|---|
| Trend pullback | For example, close above the 200-day SMA while the day’s range touches the 20-day EMA | Touch entry, recovery close or next-session order are different models |
| Consolidation breakout | For example, a close above the highest high of the preceding 20 completed sessions | Specify whether to use the next open, a limit or a stop order and when it expires |
| Range recovery | For example, an excursion below a fixed prior range followed by a close back inside | Specify the range window, tolerance and maximum recovery delay |
A candidate filter with reproducible inputs
An illustrative daily stock watchlist might require a completed close above both its 50- and 200-day SMAs and a 20-session mean of daily close × volume above $20 million. These are research choices, not universal minimums. Name the universe, feed, adjustment policy and as-of time so the same screen can be repeated.
- Compute mean daily dollar-volume proxy as mean(close × volume), not automatically mean(close) × mean(volume). Neither measures available depth at an intended order price.
- If using relative volume, state the denominator and whether the current session is complete. High aggregate volume does not identify institutional accumulation.
- Distance from a 52-week high is a price statistic. It does not establish that overhead supply is absent.
- ATR divided by price describes a historical range measure. A fixed ATR percentage band neither guarantees suitable volatility nor limits the size of tomorrow’s gap.
- Check the supported universe and fields in the selected screener. A shortlist is not proof that all listed stocks were examined.
Planned stop risk versus an overnight gap
Use invented prices: buy 100 shares at $50 with a stop reference at $48. Planned price risk is $200 and the cash position costs $5,000. If the next available execution is $44, the loss is $600 before costs—three times the planned risk. A smaller position scales this loss but does not guarantee a maximum gap.
- In the third case, the first exit is at twice the initial $2 per-share risk. Moving the remaining stop to $50 would not guarantee a $50 execution during a gap.
- FINRA distinguishes stop triggers from guaranteed fill prices. Stop-limit orders add a price condition and can fail to execute.
- Position sizes based only on stop distance omit cash or margin requirements, correlated holdings and execution capacity.
| Event | Calculation | Result before costs |
|---|
| Stop executes at planned $48 | 100 × (50 − 48) | $200 loss |
| Stop executes after a gap at $44 | 100 × (50 − 44) | $600 loss |
| Sell 50 shares at $54; remaining 50 exit at $45 | 50 × (54 − 50) + 50 × (45 − 50) | $50 net loss despite the partial profit |
The multi-session checklist
Before accepting overnight exposure, record the next known earnings release, its time zone and whether it is before or after the market session. Use the issuer’s investor-relations announcement when available; calendar estimates can change. A profit cushion does not remove event risk.
- Write a hold, reduce or exit rule before the scheduled event. Do not replace it after an unfavorable price move.
- Include dividends, splits, other corporate actions, exchange holidays and shortened sessions. Compare price series on a consistent adjustment basis.
- For a weekly filter evaluated on daily bars, say whether only the last completed week is used. A developing weekly value can change during the week.
- Specify order duration and eligible sessions with the broker. Check that exits remain active as intended after overnight maintenance or a corporate action.
- Account requirements depend on jurisdiction, broker, instrument and account type. A swing-trading label does not determine margin eligibility, settlement restrictions or a universal minimum balance.
Evaluate the holding-period tradeoff
Longer holds expose a position to more overnight events and can change capital usage, funding costs and the number of independent observations. They do not inherently produce larger net returns or less risk.
- Record holding sessions, overnight gaps, realized slippage, costs and maximum concurrent exposure alongside return and drawdown.
- Keep the candidate list as known at each historical date. A present-day list can omit delisted securities and introduce survivorship bias.
- Use an untouched evaluation period and compare fixed variants. This page has not measured a winning pattern or an optimal holding period.
- Pineify can help express supported conditions in an indicator or strategy draft. Review session, higher-timeframe and order behavior in TradingView before interpreting results.