Bat pattern — harmonic X–A–B–C–D with D at 0.886 XA

The Bat is a five-point harmonic pattern associated with Scott Carney, who dates its introduction to 2001. Its defining D level is a 0.886 retracement of XA. That geometric reference is a potential reversal zone, not an automatic entry or evidence that price will turn. B, C and the BC projection must also satisfy the selected definition.

What the Bat pattern is and where it came from

The points X–A–B–C–D describe successive swings. A bullish illustration starts with a rise from X to A and projects D inside that initial price span; the bearish construction is mirrored. The shape alone does not identify the pattern. Carney’s ratios specify the geometry, while point selection, confirmation time and trading rules require separate definitions.

The basics of trading harmonic patterns explain how to measure swings from the same chosen points. Check the rules for the specific pattern rather than assume every harmonic structure uses the same ratios or supplies a complete trading strategy.

Structure and Fibonacci levels

The formation is made of five points joined by four legs: X-A, A-B, B-C and C-D. The X-A leg is the initial reference move that defines the entire span of the pattern. The A-B leg is a pullback that ends point B below 0.618 of the XA leg, ideally at 0.382 or 0.500. This shallow pullback is the Bat's first signature, because the Gartley requires a deeper 0.618 here.

Volume One’s Bat summary gives BC/AB from 0.382 to 0.886. Check this together with D at 0.886 XA, CD/BC from 1.618 to 2.618 and CD at least equal to AB. An alternate CD = 1.27 AB is preferred rather than mandatory. The same measurement basis must be used throughout; see Fibonacci extensions when projecting targets. Do not infer how often a ratio is mismeasured or how price will react from these definitions.

Here CD = 0.01652, CD/BC ≈ 2.42941 and CD/AB = 2.065. The BC projection is within 1.618–2.618 and CD exceeds the minimum equivalent AB=CD. It is not the preferred exact 1.27 AB extension. Displaying D rounded to 1.0823 must not silently change the ratios calculated from 1.08228. D remains 0.00228, or 22.8 pips for a 0.0001 pip, above X. This is a geometric distance, not a recommended stop or maximum loss. The 0.886 ratio is 88.6% of the XA leg, not an 88.6% fall in the entire exchange rate.

How to recognise the formation step by step

Step 1 — find the X-A leg and the trend context

Fix the instrument, timeframe, price basis and rule for selecting X and A before evaluating later candles. Record when each pivot could be recognised. A higher-timeframe trend filter is an optional rule to test, not part of the ratio calculation or evidence that a Bat works best in that direction.

Step 2 — check the shallow point B and the B-C bounce

Calculate AB/XA and BC/AB from the selected points. Keep B below 0.618 distinct from the preferred 0.382 or 0.500 values. B = 0.618 fails this Bat boundary but does not by itself prove a Gartley; that pattern has additional conditions. Do not move B or C retrospectively to make a losing candidate fit.

Step 3 — confirm point D at 0.886 of the XA leg

In the bullish construction, project D = A − 0.886 × XA. Then calculate CD/BC and CD/AB from the same points. A projected D does not prove that a future swing will stop there. Minimum equivalent AB=CD and preferred 1.27 AB are different conditions. Decide measurement tolerance before inspecting the later outcome.

Entry, stop and targets — a hypothetical example

A reversal candle at D can be a condition in a proposed trading rule; it is not a required geometric measurement or a guarantee. Specify when that candle becomes available, what quote side triggers an order and when it could fill. A stop beyond X is one possible invalidation rule, not a price guaranteed by the pattern. Position size, costs, possible slippage and close-out conditions remain separate from the drawing.

If a test chooses targets at 0.382 or 0.618 of AD measured upward from D, those are that test’s exit rules. Record their exact prices and the actual entry, rather than call them inherently conservative or universally better than Gartley targets. The planned reward:risk ratio changes with confirmation delay, chosen stop, costs and fills; a smaller geometric D–X distance does not establish a higher probability of profit.

The most common mistakes when trading the Bat

1. Treating a preferred B = 0.500 as a requirement, or accepting B = 0.618 as a Bat because the remaining shape looks similar.
2. Treating a projected D as a confirmed future turning point, or using later candles to label an earlier entry.
3. Checking B and C separately without recalculating the final BC projection from that same set of points.
4. Calling a geometric D–X distance the maximum trade loss; an ordinary stop may fill worse than its trigger.
5. Changing filters or tolerances after seeing which candidates were profitable.

How the Bat differs from other harmonic patterns

The Gartley and Bat patterns have different B and D retracement requirements. The butterfly pattern and the crab pattern instead project D beyond X. An extension beyond X is a geometric distinction; it does not automatically make the formation a breakout trade or determine its return.

Even individually permitted B and C ratios can produce an invalid BC projection. In a separate fictional model, fix XA = 200 pips, AB/XA = 0.500 and AD/XA = 0.886. Then AB = 100 pips, AD = 177.2 pips and, for r = BC/AB, CD/BC = 1 + 0.772/r.

All three C retracements meet the individual C band, but the first BC projection fails. Here the upper projection limit gives r ≥ 0.772/1.618 ≈ 0.47713. This follows from our chosen B and D; it is not a new universal C boundary. The table tests ratios only, without observing market reactions or profitability.

Who this pattern is for

A reproducible rule needs point-selection and tolerance criteria as well as ratios. Historical examples chosen because they reversed do not establish how every eligible candidate performed. The geometric sources cited here do not supply an independent performance estimate for the rule you might implement; this is different from claiming that no independent study exists. Read the broader technical analysis material and Fibonacci retracements for measurement context. Any oscillator or horizontal-level filter requires its own defined comparison, not an assumption that combining tools produces an edge.

What to do tomorrow to start learning the Bat pattern

1. Fix a price series, interval and pivot-confirmation rule. Mark X–A–B–C using only information available at the stated time.
2. Calculate B, C, projected D, CD/BC and CD/AB together. Label preferred values separately from required conditions.
3. Keep invalid candidates and cases that pass through the potential reversal zone without reversing. Do not erase them after the outcome is known.
4. If testing trades, define confirmation timing, quote sides, entry and exit fills, costs and position risk separately from the pattern geometry.
5. Compare later data with frozen rules. Twenty demo attempts or an observed success rate alone does not establish live readiness, execution quality or future profitability.