Pivots and Levels

13 algorithms in Geometric Chart Patterns · 13 with asserted arithmetic.

In this family#

1. Causal Pivot Detection verified

   Finds swing highs and lows using only bars that had already arrived. Most pivot code confirms a pivot with bars that come *after* it and then plots it at the earlier index — which is lookahead bias, and it is invisible on a chart.

   detectCausalPivots(bars, leftSpan, rightSpan, minSeparation)
2. ZigZag Segmentation verified

   Segments a series into alternating swings that exceed a percentage threshold. Useful for structure, and routinely misused: the final swing is provisional and can be revised by the next bar.

   zigzagSegment(closes, threshold, minBars)
3. Support/Resistance Clustering verified

   Groups nearby pivots into levels. Clustering in basis points rather than absolute price is what makes the result meaningful across instruments — a $1 band is noise on one stock and a whole level on another.

   clusterPivotLevels(pivots, epsBps, minTouches)
4. Robust Trendline Fitting verified

   Fits a trendline through pivots in log space with outlier resistance. Log space matters: a straight line in price implies a constant *dollar* change per bar, which is not what a trend on a long chart means.

   fitRobustTrendline(pivots, kind, toleranceBps, minInliers, maxPoints, projectionIndex)
5. Classic and Floor-Trader Pivot Points verified

   Projects the classic floor-trader pivot and two resistance/support pairs for each bar from the previous bar's high, low, and close. The pivot is (H + L + C) / 3, r1/s1 reflect it across the prior low and high, and r2/s2 add and subtract the prior bar's range.

   classicAndFloorTraderPivotPoints(input)
6. Fibonacci Pivot Points verified

   Takes the same (H + L + C) / 3 pivot as the classic variant but places the bands at Fibonacci fractions of the previous bar's range: r1/s1 at 0.382 of it and r2/s2 at 0.618.

   fibonacciPivotPoints(input)
7. Camarilla Pivot Points verified

   Anchors its bands on the previous bar's close rather than on the pivot: r1/s1 sit at close ± 1.1 × range / 12 and r2/s2 at close ± 1.1 × range / 6. The pivot series is still the conventional (H + L + C) / 3, which is why the bands are not symmetric around it.

   camarillaPivotPoints(input)
8. Woodie Pivot Points verified

   Weights the current bar's open into the pivot, (H_prev + L_prev + 2 × open) / 4, then derives r1/s1 by reflecting that pivot across the previous low and high and r2/s2 by adding and subtracting the previous bar's range.

   woodiePivotPoints(input)
9. DeMark Pivot Points verified

   Chooses one of three formulas for the intermediate X depending on whether the previous bar closed below, above, or level with its open, then reports pivot = X / 4 with a single band pair r1 = X / 2 - L_prev and s1 = X / 2 - H_prev.

   demarkPivotPoints(input)
10. Rolling Support and Resistance verified

    Reports, for each bar, the lowest low and highest high of the period bars that came strictly before it. Excluding the current bar is what makes the level a thing price can break rather than a level price defines.

    rollingSupportAndResistance(input)
11. Fractal Support and Resistance verified

    Holds the most recent confirmed five-bar fractal high and low forward as resistance and support. This deliberately delegates to the same causal five-bar confirmation used by Fractal Chaos Bands, so a fractal at bar p is only reported from bar p+2 onward.

    fractalSupportAndResistance(input)
12. ATR Support and Resistance verified

    Places support and resistance a volatility-scaled distance either side of a simple moving average of close: center ± multiplier × ATR(period), where the ATR is Wilder-smoothed true range.

    atrSupportAndResistance(input)
13. Pivot Range Width verified

    Measures the central pivot range rather than its bands: bottom is the previous bar's midpoint (H + L) / 2, top is that midpoint reflected through the pivot, and width is the absolute distance between them.

    pivotRangeWidth(input)

What they share#

Every topic here is a row-classify</code> or <code>record-transform, so once you have called one the rest follow the same shape. Import paths differ only in the final segment:

import { detectCausalPivots } from "fintech-algorithms/geometric-chart-patterns/pivots-and-levels/causal-pivot-detection";
import { zigzagSegment } from "fintech-algorithms/geometric-chart-patterns/pivots-and-levels/zigzag-segmentation";

Read them in the order above — the sequence is pedagogical, not alphabetical.

Where this sits#

Geometric Chart Patterns collects 64 algorithms across 7 families. For the concept behind this family rather than the call signatures, see the concept guides.