Is the bump and run pattern bullish or bearish?

Both variants exist. This example is a bearish top. Bulkowski describes a bottom with a descending lead-in line, a deeper downward bump and confirmation by a close above the original line. A bottom needs its own high-anchor and distance rules; changing only the label on this top example is insufficient.

Does the bump have to be twice the lead-in height?

Two times is a commonly described geometric guideline, not a measured probability. This page uses the highest high in each selected phase and subtracts the original line value at that same bar. Its hypothetical ratio is 16 / 4.80 = 3.33. Record the measurement convention because source descriptions differ.

Can I identify the pattern from a 45-degree line?

A screen angle alone cannot define it. Axis scaling and chart dimensions change the angle. Save the coordinates and calculate slopes in price units per bar, then apply a stated distance and confirmation rule. Logarithmic distances require a separate specification.

What invalidates a bump and run candidate?

The answer depends on the frozen rules. This example rejects a new high above 128 before confirmation and expires after bar 42 if no closing break occurs. After confirmation, a close above the rising line plus 0.25 within three bars labels a failed run. These are explicit review choices, not universal thresholds.

What is the bump and run success rate?

This page establishes no success rate. A usable estimate needs a fixed detector, instrument universe, timeframe, sample period and definition of success, plus trading costs if profit is the outcome. The invented diagram verifies arithmetic and event classification only.

TradingView chart patterns

Bump and run pattern: measure the acceleration and the break

A bump and run pattern is a potential reversal with three stages: an initial trend, a steeper move away from its trendline, then a reversal through that original line. In a bearish top, the lead-in rises, the bump accelerates upward and the run begins with a break below lead-in support. A sharp rise alone is insufficient. This guide requires a completed closing break and distinguishes it from a wick, an expired candidate and a failed reversal.

See the decision guide

Lead-in, bump and run describe a sequence

StockCharts describes a moderate advance, a faster rise away from its baseline, and a return that eventually breaks that baseline. Its presentation uses roughly 30 to 45 degrees for the lead-in and 45 to 60 degrees for the bump, while acknowledging that angles depend on chart scale and size. It compares the bump peak with the lead-in high using vertical distance to the original line; the bump distance should be at least twice as large.

Bulkowski specifies an arithmetic chart and a close below the rising lead-in line for a confirmed top. His current lead-in-height description uses the widest distance in the first quarter and references a price low. StockCharts uses the highest lead-in high. Record the convention: these descriptions are not identical.
The worked example below uses the StockCharts high-to-line convention with an explicitly selected lead-in window. It does not reproduce every rule in either source.
A bubble or speculation narrative cannot be inferred from this geometry alone. The observable claim is that price accelerated relative to a chosen baseline.

Stage	Observable structure	What is still missing
Lead-in	A rising support line through separated lows	The later acceleration and reversal
Bump	A steeper ascent with a larger distance above that line	A completed break of the original support
Run	A closing break below the original lead-in line	Evidence that the decline will persist

An original example with checkable coordinates

This invented arithmetic-scale path shows selected observations, not every candle or a historical trade. Bar numbers refer to one fixed timeframe. At the close of bar 20, save lead-in lows A = (0, 100), B = (10, 104) and C = (20, 108). The line is L(t) = 100 + 0.40t. Assume all lead-in lows stay on or above L(t) minus 0.20 and its highest high is 112 at bar 18. Freeze these anchors before examining the bump.

Hypothetical bump and run top on an arithmetic price scale. Lead-in support connects bars 0, 10 and 20; a peak of 128 at bar 30 stands 16 above the line. A closing break occurs at bar 38. — Original teaching diagram with invented prices. Selected highs, lows and closes are joined for orientation; the joins are not a complete candle series. The dashed branch is an alternative failed-break path. View original source

For this review, require positive lead-in slope, three low touches within 0.20 of the line, and a bump slope at least twice the lead-in slope. D = (22, 110) and E = (28, 122) are manually selected low observations. A slope ratio of two is a custom filter, not the same as a 50% increase in screen angle.
At the close of bar 32, assume the highest high since bar 21 is 128 at bar 30, followed by closes of 126 and 124. Save that peak and begin the confirmation window at bar 33. Assume no earlier completed close has broken below L(t). The two declining closes are the example rollover rule.
This is a review of one preselected candidate. An automated scanner also needs a start-bar rule, a reproducible anchor-selection method and an overlap policy. If anchors use future-bar pivots, their availability time must be recorded separately from their plotted time.

Measurement	Calculation	Result
Lead-in slope	(108 - 100) / (20 - 0)	0.40 price units per bar
Lead-in height H	112 - L(18) = 112 - 107.20	4.80 price units
Bump slope between selected lows D and E	(122 - 110) / (28 - 22)	2.00 price units per bar
Slope ratio	2.00 / 0.40	5.00
Bump height B at bar 30	128 - L(30) = 128 - 112	16.00 price units
Height ratio B / H	16 / 4.80	3.33, above the illustrative minimum of 2

Confirm the run without moving the baseline

For the frozen candidate, evaluate completed bars 33 through 42. First reject a bar whose high exceeds the saved 128 peak. Otherwise confirm on the first close strictly below L(t) - 0.25. A touch or equality does not qualify. The fixed 0.25 price-unit buffer and ten-bar window are teaching choices, not recommended market settings. Assume intervening closes do not qualify and intervening highs remain at or below 128.

The same 114.80 close has different classifications at bars 37 and 38 because the baseline rises. Extending a frozen line means keeping its slope and anchors fixed, not keeping its price constant.
A close below the steeper bump line only says the acceleration has weakened. It does not meet the original lead-in-line break rule.
Rejecting a new high is specific to this saved single-bump candidate. Broader pattern descriptions can admit multiple bumps; record a new candidate separately rather than rewriting the old one.

Observation	Threshold at that bar	Classification
Bar 36: low 113.80, close 114.60	L(36) = 114.40; closing threshold 114.15	Wick below support; unconfirmed
Bar 37: close 114.80	L(37) = 114.80; closing threshold 114.55	Close on support; unconfirmed
Bar 38: close 114.80	L(38) = 115.20; closing threshold 114.95	First qualifying close: confirmed run
Alternative bar 38: high 128.10, close 114.80	Saved peak is 128	Rejected first, even though the close breaks support
Alternative: no qualifying close through bar 42	Window ends at the close of bar 42	Expired without confirmation

Separate a failed break from an unconfirmed pattern

For this example, label a confirmed run failed if any of the next three completed closes is strictly above L(t) + 0.25. After confirmation at bar 38, a bar 40 close of 116.50 exceeds 116.25 and meets that failure rule. The earlier confirmation remains in the record. A return to the line that closes below the buffer does not meet this custom failure definition.

A new peak before confirmation rejects this candidate; a recovery after confirmation is a failed-break outcome. Those events answer different questions and should not be pooled.
A failure label is not a protective stop. Entry timing, executable fills, position size, stop orders, holding period, fees and slippage need separate rules before evaluating a strategy.
Keep candidates that expire or fail in the review. A collection of diagrams chosen after large declines cannot establish a success rate. No performance estimate or price target is calculated here.

Why screen angles and scale changes alter the test

TradingView distinguishes Regular scaling, which displays absolute price differences, from Logarithmic scaling, which displays proportional changes. It also provides a price-to-bar ratio setting. On a linear screen, a line angle obeys tan(angle) = slope × pixels per price unit / pixels per bar. Resizing either axis changes that angle without changing the price observations.

The numeric slope ratio compares both segments in the same price units and timeframe. It survives uniform linear axis resizing. It does not make a five-minute sample interchangeable with a daily sample, and it becomes unstable when the lead-in slope is near zero. Reject a flat baseline; choose any minimum normalized slope before testing.
The two-to-one height check here uses vertical arithmetic differences. On a positive-price log chart, the corresponding distances are ln(high / baseline). That is a different measurement. Recompute and label it separately; do not measure one height in pixels and the other in dollars.
Keep the symbol, timeframe, session, price adjustments and chart type fixed during the review. A 0.20 touch tolerance or 0.25 buffer has no universal meaning across differently priced instruments. Percent or volatility normalization would require a new, stated rule.
StockCharts discusses volume expansion during the acceleration. This example has no volume data and applies no volume filter. Adding one requires its baseline, threshold and missing-data behavior to be specified first.

Use the annotated chart as a review input

In TradingView, retain enough history to show the lead-in, label the selected low coordinates and display the price and time axes. Save the chart scale and the close used for confirmation. Upload that screenshot to Pineify AI Chart Analysis to ask for a second reading of the candidate structure. Check any proposed coordinates against the chart; the screenshot does not establish exact OHLC values or future results.

For a Pine Script draft, give the Coding Agent the anchor rule, high-to-line height convention, touch tolerance, slope ratio, confirmation buffer, expiry and failure horizon. Ask it to expose the observation time of each signal and verify the generated source.
Review event timestamps against this table before adding trading orders. Syntax validation checks code acceptance; it does not prove that the detector matches the intended geometry or has trading value.

Where Pineify fits

AI Chart Analysis

Upload a screenshot with symbol, timeframe, scale and marked anchors to review the candidate structure. Verify the returned interpretation against the source chart.

Also useful: Pine Script AI Coding Agent. Turn explicit anchor, distance, tolerance and closing rules into a script draft for inspection and syntax checking. This page does not provide an automatic pattern scanner.

Review an annotated chart

Related resources

TradingView chart patterns Bull flag: a continuation pause Double bottom: neckline confirmation Backtesting accuracy and execution assumptions

Frequently asked questions

Educational information, not investment advice. All diagram prices, thresholds and outcomes are hypothetical. No backtest, recognition accuracy or expected return is presented. A confirmed pattern can fail and does not guarantee an execution price or profit.

Sources and verification

Last verified: 2026-10-17

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.