In a hypothetical EUR/USD bar, price reaches Low 1.0820 and later closes at 1.0860 near High. That is a description of observations, not evidence that identifiable buyers defended support or that the next bar will rise. A pin bar’s geometry, its location and a trade’s execution are separate questions.

What a pin bar is and how to spot one

A pin bar is a candle with a small body near one end and a dominant shadow on the other. A lower shadow shows that Close ended above Low; an upper shadow shows Close below High. OHLC does not identify the participants or their motives. The same shape can be studied under different names and preceding-trend conditions; see which candlestick patterns are worth knowing first.

Chosen geometric rules for this exercise
Wick lengthdominant shadow ≥ 2B; opposite shadow ≤ 0.10R
Body positionwhole body in the upper third for a lower wick, lower third for an upper wick
Closing priceR = High − Low; B = |Close − Open|; 0 < B ≤ R/3
SilhouetteGeometry only; zero-body doji excluded and equality allowed

Bullish versus bearish pin bar — who rejected whom

A lower-wick candidate has its small body near the top; an upper-wick candidate near the bottom. The body can be rising or falling. A hammer and a hanging man both have a long lower shadow, distinguished by the preceding decline or rise. A shooting star has a long upper shadow after a rise; the hanging man is not its mirror. Compare the inverted hammer and the shooting star.

The shape is unchanged, but the filter fails

Shifting all four prices equally preserves candle proportions. It does not preserve their relationship to a previously fixed level.

Scroll sideways to see the full diagram

A shifted pin bar keeps its proportionsTwo identically proportioned candles are shifted by four points. The first Low lies inside the fixed 99–101 zone; the second Low 104 lies above it.SAME GEOMETRY, DIFFERENT LOCATION100104108112A: Low 100B: Low 104Touches zoneOutside zonePreviously fixed zone:99–101.B is A shifted by+4 points.Both: range 6; body 0.5; lower wick 5.Only A passes the location filter.

Fictional prices in arbitrary points, identical scale and quote side. Zone 99–101 was fixed before evaluating both separate variants and stays unchanged. A: O105, H106, L100, C105.5. B: O109, H110, L104, C109.5. Both meet previously specified lower-pin-bar proportions; the separate original filter requires Low inside the zone. The document CME Group: chart types supports the distinction between the Open–Close body and shadows extending to High and Low.

The gold band is shared by both variants, not fitted to each wick.

Both have B/R=0.5/6, D/B=5/0.5=10 and U/R=0.5/6, so shape classification matches. A has Low 100 inside the zone; B has Low 104 outside. Rejecting B follows the recorded location rule; it does not prove price will rise after A.

Test geometry and location separately

Define the context before seeing later prices. For example, a previously fixed support zone is an additional condition; it does not alter the OHLC proportions or demonstrate an edge. Specify how you draw support and resistance and when a higher-timeframe observation becomes available, rather than fitting the context to a successful outcome.

Hypothetical EUR/USD candles: Bid, 1 pip = 0.0001
Lower-wick example: O/H/L/C1.1050 / 1.1060 / 1.1000 / 1.1055
Upper-wick example: O/H/L/C1.1005 / 1.1060 / 1.1000 / 1.1010
Both examplesrange 60 pips, rising body 5 pips, dominant wick 50 pips
Separate location conditionfirst Low 1.1000 is inside a previously fixed 1.0995–1.1005 zone

A pattern's shape alone does not establish a hit rate or an edge on forex. Any published stock-market candle statistics depend on the sample and success definition and cannot be transferred directly to a currency pair. Test the chosen location, entry and exit rules together, including costs.

Measurements: let U = High − max(Open, Close), D = min(Open, Close) − Low. The lower-wick example has D = 50 and U = 5 pips; the upper-wick example D = 5 and U = 50 pips. Both bodies occupy about 8.33% of range, and both dominant wicks 83.33%. For the lower variant require min(Open, Close) ≥ Low + 2R/3; for the upper max(Open, Close) ≤ Low + R/3. Both satisfy our model, despite sharing a rising body.

Bid/Ask counterexample: take all four successive Bid observations [1.1050, 1.1060, 1.1000, 1.1055]. Adding a constant 2-pip spread preserves the shape. Instead use spreads [2, 2, 40, 20] pips at those observations: Ask becomes [1.1052, 1.1062, 1.1040, 1.1075]. Its own OHLC is 1.1052/1.1075/1.1040/1.1075: range 35 pips, body 23, lower wick 12 and upper wick zero. Its body exceeds R/3, so Ask fails our shape test while Bid passes. These deliberately large spreads are a counterexample, not typical EUR/USD conditions; a variable-spread Ask candle cannot be rebuilt by adding one spread to finished Bid OHLC.

How to trade the pin bar step by step

Recognition does not select an entry. Buying after the close, awaiting a High breakout and placing a pullback limit are different rules with different possible fills. A limit may remain unfilled. A breakout shows that the chosen level was crossed; it cannot confirm that the direction will persist. Record which observations and prices were available when each order could be sent.

For an original model, buy 10,000 EUR after the first candle completes, at Ask 1.1058 with Bid 1.1056, using a USD account. Choose stop 1.0998 (2 pips below the earlier Low) and target 1.1178. A close at Bid 1.0998 loses 60 pips/USD 60 gross; at Bid 1.1178 it gains 120 pips/USD 120. With USD 2 total commission the alternatives are −USD 62 and +USD 118: gross reward/risk 2:1 becomes 118/62 ≈ 1.90:1. Spread is already reflected in Ask entry and Bid exit, so do not subtract it again. Financing, taxes and currency conversion are omitted.

An ordinary stop need not fill at its level. An alternative Bid exit at 1.0993 loses USD 65 gross and USD 67 after commission. The 5-pip adverse fill is hypothetical, not a frequency estimate. For typical MT5 forex, long stops are checked on Bid and short stops on Ask; exchange instruments may use different trigger rules. If a candle includes both stop and target, OHLC alone may not establish which came first.

The most common pin bar trading mistakes

Common testing errors confuse a shape condition with a trading rule or use information that was unavailable at the decision time.

  • Assuming a hit rate from shape. A candle definition gives a classification, not the probability or net outcome of a trade.
  • Using a final shape before the close. Live OHLC can change. A completed-candle rule and an intrabar rule are different tests.
  • Choosing a stop only to reduce the displayed risk. Its distance, trigger side, size and possible fill must belong to the same defined rule. No universal buffer follows from pin-bar geometry.
  • Changing context after the outcome. Define any preceding-trend filter before evaluation. Apply the same principle to an engulfing pattern; neither shape establishes a preferred context or advantage by itself.
  • Assuming a timeframe ranking. H4, D1 and M15 aggregate different observations. Compare a complete rule and costs instead of assuming lower bars have no information.

What to do tomorrow

  1. Measure the chosen shapes. Fix the OHLC validity, equality and zero-body rules. Record the instrument, quote side, timeframe and feed before evaluating context.
  2. Write your entry, stop and target rules before the test. For a bullish pin bar place the invalidation stop beyond the lower wick; for a bearish one, beyond the upper wick. Define the buffer and required reward-to-risk ratio for the instrument, and record costs and possible slippage.
  3. Test the context separately. Keep zones fixed before each signal. Two conditions need not be statistically independent, and their combination does not establish an A-grade outcome.
  4. Keep evaluation distinct from fitting. Log filled, cancelled and unfilled orders as well as losses. Twenty demo trades do not establish effectiveness or automatically justify risking live capital.

If you want to organise the broader picture of price action within a structured reference, the technical analysis section on ForexMechanics.com is a good place to continue.