Pineify logo

TradingView strategy research

Silver Bullet strategy: session windows and testable rules

The ICT Silver Bullet strategy is a time-window trading model associated with Michael J. Huddleston, The Inner Circle Trader. Its named windows are 03:00 to 04:00, 10:00 to 11:00, and 14:00 to 15:00 in New York local time. A window identifies when to look for a setup; an FVG and a later entry trigger still need separate rules. This page defines one untested morning variation for auditing on TradingView.

See the decision guide

Source and scope of this rulebook

The original teaching reference is the 2023 ICT Mentorship lesson, ICT Silver Bullet Time Based Trading Model, linked below. It places fair value gaps in a time window and a broader directional liquidity context. The clock alone does not define a trade. The mechanical sweep reference, displacement threshold, midpoint order, and noon exit below are choices for this research variation, not a claim to reproduce every discretionary decision in that lesson.

  • A move through a prior high or low is observable in candles. Calling that level liquidity does not prove that actual stop orders were resting there.
  • Use the linked ICT overview for the broader vocabulary and the FVG guide for gap identification. This page focuses on the sequence and timing of one setup.

Silver Bullet times and daylight saving

The named sessions use New York local time, including the window called London. They are not three hours measured on a London clock. The UTC columns below are conversions, not alternative session definitions.

  • TradingView distinguishes chart display time from script calculation time. Changing the displayed zone does not change a script; time functions default to the exchange zone unless another zone is specified.
  • America/New_York follows daylight saving changes. UTC-5 stays fixed. For example, 10:30 New York is 15:30 UTC on January 15, 2026, but 14:30 UTC on July 15, 2026.
  • Treat the start as inclusive and the end as exclusive for new entries in this test. At 11:00, cancel unfilled morning orders. An existing position follows its separate exit policy.
WindowNew YorkUTC during ESTUTC during EDT
London03:00 to 04:0008:00 to 09:0007:00 to 08:00
New York morning10:00 to 11:0015:00 to 16:0014:00 to 15:00
New York afternoon14:00 to 15:0019:00 to 20:0018:00 to 19:00

One explicit morning setup to test

Use standard one-minute candles on one symbol and feed, Monday through Friday in New York time. Start flat. Freeze the high H and low L of the 30 completed bars opening from 09:30 through 09:59. Skip days with missing reference bars or H equal to L. These reference levels and all thresholds below are illustrative research settings.

  • All three FVG bars must open at or after 10:00 and finish before 11:00. Require a positive bar 2 range. A tied gap boundary fails. Round a long midpoint down and a short midpoint up to a valid tick.
  • Accept the first sweep direction only. Skip the day if the sweep bar also crosses the opposite reference level. Before the FVG is confirmed, a touch of S or of the target after the sweep cancels the candidate.
  • Reject an FVG whose confirmation bar has already reached the target. Once submitted, keep the order prices fixed. The stop and target must activate with a fill; do not add an unrecorded close-based cancellation to a resting order.
  • Allow one submitted entry order per day, no replacement and no pyramiding. Use one tradable unit with sufficient capital as a test convention. If the limit never fills before 11:00, record an expired order.
  • An open position keeps its stop and target after 11:00. If still open at noon, submit a market exit and record its actual simulated fill. This noon deadline is a custom holding-period rule.
StageLong variationShort variation
Sweep and reclaimAfter 10:00, the first bar with low < L and close > L arms a long candidate. Save its low S.The first bar with high > H and close < H arms a short candidate. Save its high S.
Displacement and FVGTake the first qualifying sequence of three consecutive completed bars after the sweep bar. Bar 3 low must exceed bar 1 high.Bar 3 high must be below bar 1 low, using the same sequence and timing rules.
Directional confirmationBar 2 closes above its open, its body is at least half its range, and bar 3 closes above the sweep bar high.Bar 2 closes below its open, its body is at least half its range, and bar 3 closes below the sweep bar low.
Order creationAfter bar 3 closes, submit a buy limit at the midpoint of bar 1 high and bar 3 low.Submit a sell limit at the midpoint of bar 1 low and bar 3 high.
Frozen exitsStop one tick below S; target H. Require stop < limit < target.Stop one tick above S; target L. Require target < limit < stop.

A valid sequence and three rejected alternatives

Assume fictional prices, a 0.25 tick, L = 100 and H = 104. All times below are New York bar-opening times on one-minute candles. These values demonstrate the rulebook; they are not a recorded trade or evidence of performance.

  • A valid 10:12 entry may still hit the stop. Passing the sequence checks says nothing about the probability of a winning trade.
  • Using a fixed 15:00 to 16:00 UTC morning window in July would scan 11:00 to 12:00 New York. That is a different experiment, even if the chart looks similar.
Time or caseKnown pricesDecision
10:04 sweep barHigh 100.75, low 99.50, close 100.25.It reclaims L. Save S = 99.50 and wait.
10:05, 10:06, 10:07 FVG barsBar 1 high 100.75; bar 2 O/H/L/C = 100.50/102.00/100.50/101.75; bar 3 low 101.25 and close 101.75.Assume no intervening invalidation. At 10:08, gap 100.75 to 101.25 and directional checks pass.
After 10:08 confirmationLimit = (100.75 + 101.25) / 2 = 101.00; stop = 99.25; target = 104.00.Submit the limit. A later executable retracement is required for a fill.
Hypothetical 10:12 fillAssume a fill at 101.00 with no earlier exit condition.Planned stop distance is 1.75; target distance is 3.00. Costs and gaps change realized outcomes.
Overlapping candlesReplace bar 3 low with 100.50.Reject: 100.50 is below bar 1 high. There is no bullish FVG.
Late confirmationThe same FVG first becomes known at 11:00.Reject: the completed setup is outside the entry deadline.
Late retracementThe valid 10:08 order first reaches its limit at 11:02.No entry: the pending order expired at 11:00.

Audit the backtest before interpreting results

Keep the window definition fixed while evaluating this variation. Record symbol, feed, candle type, date range, session coverage, tick size, point value, commission, spread treatment and slippage. Reserve later dates before choosing any parameter. Compare the same written rules on the earlier and later samples.

  • Log sweep detection, FVG confirmation, order submission, fill, cancellation and exit timestamps. A box drawn on earlier bars must not authorize an earlier order.
  • Inspect the 11:00 boundary explicitly. Historical bar calculations and resting limit orders can produce an unwanted boundary fill unless cancellation timing is implemented and tested.
  • TradingView uses a broker emulator with order timing and intrabar assumptions. A candle touching a limit is not a guarantee of a live fill. Inspect bars containing both entry and stop, or both stop and target, with finer data where available.
  • Separate no setup, invalid setup, expired order and filled trade counts. Report net results, drawdown and sample size; this page supplies no measured win rate or return.
  • Choose any scheduled-news exclusion before testing and retain the calendar used. Do not remove losing event days after seeing results.

Give Pineify the complete specification

Ask Pine Script AI Coding Agent to draft this morning variation with the named time zone, frozen reference range, ordered setup stages, entry expiry and separate noon exit. Request visible event timestamps and compare them with the valid and rejected examples before evaluating a backtest. Review generated source and syntax results; compilation alone does not establish faithful execution.

  • After the baseline matches the specification, use Strategy Optimizer to compare a small, predeclared set of exposed inputs, such as displacement body fractions of 0.50 and 0.60. Keep the time zone, sample and costs identical.
  • Keep the later evaluation period separate from parameter selection. Optimizer output cannot validate the original ICT method or turn this illustrative rulebook into a verified trading edge.
Where Pineify fits

Pine Script AI Coding Agent

Draft a strategy from the explicit session and order rules, then inspect its source and timestamped behavior against the worked cases.

Also useful: Strategy Optimizer. Compare limited exposed parameter choices after validating the baseline. Preserve a separate evaluation period and account for trading costs.

Draft the session-based strategy

Related resources

TradingView strategy guidesICT strategy conceptsFair value gap identificationBacktest accuracy checks

Frequently asked questions

Educational strategy research, not investment advice. The mechanical variation and prices are illustrative and have not been performance-tested. Trading can lose money; time filters, backtests and parameter searches do 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.