Pineify logo

TradingView strategy research

Elder Triple Screen on TradingView: timeframes and trigger rules

The Triple Screen trading system separates three decisions: choose a direction on a longer timeframe, find a countertrend move on an intermediate timeframe, then define an entry trigger. Alexander Elder describes this sequence in Come Into My Trading Room. Three indicators agreeing on one candle do not reproduce that sequence. On TradingView, each decision also needs a timestamp showing when its inputs became available.

See the decision guide

What belongs to Elder, and what this test changes

In the Triple Screen Update chapter, Elder assigns trend analysis to the first screen, oscillator-based timing to the second, and order placement to the third. He discusses approximately fivefold spacing between timeframes, with rounding, rather than a universal set of chart intervals. The third screen can use intermediate or shorter charts; a third indicator is not required. His original trend filter used weekly MACD-Histogram slope; the later book also discusses a weekly EMA. A two-day smoothed Force Index is one pullback measure.

  • The book describes several entry techniques, including intraday price breaks. The example here deliberately waits for an hourly close before submitting a market order. That delay changes both eligible entries and fill prices.
  • Elder Impulse and Force Index answer narrower indicator questions. This page concerns the order of decisions, the availability of their data, and how a setup becomes a trade.
  • The book reference below identifies the author's edition. Consult the Triple Screen Update chapter, pages 127 to 140. Elder.com also describes the longer-chart strategic decision and shorter-chart entry decision in its TradeStation manual.

Freeze one weekly, daily and hourly rule set

Use standard candles for one symbol, a fixed exchange session and unchanged adjustment settings. Run the test on a 60-minute chart. The daily and weekly series must use that same symbol and session context. This baseline takes one share, one position at a time, with sufficient capital and no pyramiding. It trades only long; a bearish weekly filter means stand aside.

  • Require at least 100 completed weekly bars and 30 completed daily bars before evaluating entries. These are declared warm-up choices, not proof that every feed produces identical EMA seeds.
  • At each new session, replace the previous setup with the latest completed daily test and levels. The current daily Force Index may turn positive intraday without cancelling this frozen session setup.
  • Weekly histogram direction depends on its slope, not its sign. A move from -0.80 to -0.50 passes the rising test; a move from 0.80 to 0.50 fails it.
ScreenFrozen research ruleAvailable information
Weekly directionCompute MACD as closing-price EMA12 minus EMA26, its signal as EMA9 of MACD, and histogram as MACD minus signal. Permit longs only if the latest completed weekly histogram exceeds the preceding completed value.Use the two completed weeks, never the developing week. Equality means no long setup.
Daily pullbackRaw Force Index is daily volume multiplied by close minus previous close. Require its two-day EMA to be strictly negative on the last completed daily bar.At the next session, freeze that daily high H and low L. Missing volume, indicator values or history means no setup.
Hourly triggerWhile the weekly and daily tests pass, require a completed hourly close at or above H plus one minimum tick. Submit a market entry after that close.A wick above the level is insufficient. Only the first eligible trigger in the session can enter.

Make cancellation and exits part of the specification

Set the invalidation and protective stop S to L minus one minimum tick. Before entry, any hourly low at or below S cancels the setup for that session. Check this condition before the close trigger, including when both occur in the same bar. A cancelled setup cannot re-arm until a later session.

  • Expire an unfilled setup at the session end. Do not accept an entry trigger on the final session bar, including a shortened last bar. This prevents the baseline from intentionally scheduling a new entry for the next session open.
  • Let E be the accepted trigger close. Freeze a target T = E + 2 × (E - S), requiring E > S. Submit the entry-linked protective stop and target with the entry order; keep both prices fixed. The target is based on the signal, not on an unknown future fill.
  • Count the entry bar as zero. If neither exit has filled by the close of hourly bar ten after entry, submit a market exit and retain the bracket until flat. Count only available session bars; an open position can remain overnight. Ignore new setups while an entry is pending or a position is open, and allow no re-entry in the same session.
  • A later weekly filter change blocks new entries but does not replace these exits. If the next opening price is beyond S or T, retain the actual entry and executable exit in the simulation rather than deleting the trade after learning its fill.

Align the higher-timeframe clock before testing signals

TradingView documents that unconfirmed higher-timeframe values can change, and that requesting final values too early introduces lookahead bias. Its confirmed-data approach offsets the expression inside the higher-timeframe request and uses the appropriate lookahead setting. Offset the weekly or daily expression in its own context, not the returned hourly series. The former means a previous weekly or daily bar; the latter means one hourly observation.

  • Calculate each indicator on its native timeframe. Applying MACD to a weekly close repeated on every hourly bar produces a different series.
  • Use exchange sessions and actual bar timestamps. Holidays, shortened sessions and overnight markets invalidate assumptions such as exactly five daily bars per week or exactly six hourly bars per day.
  • Record the source daily and weekly close times beside every trigger. Compare historical and replayed signals after reload. A clean syntax check cannot prove that these timestamps are correct.
Decision pointPermitted inputInput that must be excluded
Tuesday hourly closeWeekly filter from the last two completed weeks; Monday daily setup.The eventual histogram of the week containing Tuesday, or Tuesday final daily volume.
First session of a new weekRefresh the weekly filter from the week that has just completed; use the previous completed session for the daily setup.A filter chosen from the future end of the new week.
Friday before the weekly closeContinue using the previously confirmed weekly pair under this baseline.Promoting an intraday Friday histogram to a completed weekly result.

Worked example: a valid setup and two rejected triggers

All figures are hypothetical. Assume the latest completed weekly histograms are -0.80 and then -0.50, Monday Force Index EMA2 is -1,200, Monday high H is 101.20, low L is 99.60, and the minimum tick is 0.01. On Tuesday the weekly filter passes, the daily pullback qualifies, the trigger threshold is 101.21 and S is 99.59.

  • With default bar-close strategy calculation, TradingView normally fills a newly submitted market order at the next bar open. The assumed 101.30 is not a promised fill.
  • A later current-week histogram of -0.30 cannot be used to justify Tuesday retrospectively. If the previously completed weekly pair had been falling instead, Tuesday would have no eligible setup.
Tuesday observationCalculation or decisionResult
First hour: high 101.35, low 100.20, close 101.15101.15 < 101.21 despite the higher wick.No entry.
Second hour: high 101.40, low 100.80, close 101.25No invalidation; 101.25 ≥ 101.21. E = 101.25.Submit entry and bracket after this close.
Frozen targetSignal distance = 101.25 - 99.59 = 1.66; T = 101.25 + 2 × 1.66 = 104.57.The signal-based target distance is 3.32.
Assumed next-bar fill: 101.30Actual stop distance = 1.71; target distance = 3.27; ratio ≈ 1.91.The fill does not preserve a 2:1 payoff before costs.
Separate counterexample: second-hour low 99.58, close 101.2599.58 ≤ 99.59 invalidates before the trigger check.Cancel for Tuesday, even though the close clears 101.21.

Test the sequence before optimizing its parameters

Ask Pine Script AI Coding Agent to implement the frozen specification with source timestamps, setup state, cancellation reason and entry markers. Inspect the generated source, run its syntax check, and compare the event sequence with the numerical example. This page supplies a rule specification, not an executed strategy or backtest.

  • After timing agrees, Strategy Optimizer can compare exposed strategy inputs. For example, predeclare Force Index lengths 2 and 3 and time exits 8, 10 and 12 bars: 2 × 3 = 6 combinations. Keep the other rules fixed and save the settings used for each run.
  • Choose a development interval and a later untouched interval before comparing results. Select parameters only on the development interval; run the frozen selection separately on the later dates. This workflow does not claim automatic walk-forward validation.
  • Export supported TradingView trade data to Backtest Deep Report to inspect trade count, profit factor and drawdown from its reconstructed closed-trade equity curve. That curve cannot establish the largest intratrade equity loss or validate higher-timeframe alignment in the script.
  • Include commission and slippage; inspect gaps and bars that touch both exits. TradingView historical fills depend on broker-emulator assumptions and available intrabar data. Small trade samples and repeated parameter selection can make an apparent winner unreliable.
  • A daily pullback can become a reversal, and a weekly filter can react after substantial price movement. Log cancelled setups and no-trade periods as well as fills. None of the three screens guarantees continuation or live execution.
Where Pineify fits

Pine Script AI Coding Agent

Draft the specified weekly filter, frozen daily setup and hourly trigger. Review source timestamps and cancellation behavior before interpreting results.

Also useful: Strategy Optimizer. Compare explicit strategy input combinations after the baseline passes timing checks. Keep a later evaluation interval separate from parameter selection.

Draft the Triple Screen rules

Related resources

TradingView strategy guidesDefine a pullback and its invalidationAudit backtest accuracyReview exported trades

Frequently asked questions

Educational strategy specification, not investment advice. The custom rules and hypothetical prices have not been performance-tested. Trading can lose money; gaps and costs can increase losses beyond a planned stop distance. Historical simulation does not guarantee future results.

Sources and verification

Published by Pineify. Pineify is an independent product and is not affiliated with TradingView. Sources support the technical explanations; examples do not establish trading performance.