# Andrews' Pitchfork

Also known as: Schiff, modified Schiff, median line, warning lines, inside/outside rules.
A Trend concept (Channels, lines & geometry) in the LuxAlgo Library, with 1 indicator implementation.

## What is Andrews' Pitchfork?

Andrews' Pitchfork is a three-point channel tool built around a median line. From a pivot A at the start of a move and the next [swing high and low](https://www.luxalgo.com/library/concept/swing-high-low/) B and C, the median line runs from A through the midpoint of B-C; two parallels drawn through B and C complete the fork. Alan Andrews' method treats the median line as the trend's center of gravity: price is expected to gravitate back toward it, while the outer tines act like the walls of a [parallel channel](https://www.luxalgo.com/library/concept/parallel-channel/).

The tool carries its inventor's name. Alan Hall Andrews, an engineer by training, taught the technique in his Action-Reaction Course in the mid-20th century, building on the older action-reaction idea that market swings tend to answer each other symmetrically around a central tendency. His claim was never that price obeys the fork, but that the median line gives a disciplined way to define the trend's center so that behavior around it becomes readable.

That expectation is a tendency, not a rule. Plenty of trends never return to the median line, and in Andrews' own teaching (the Hagopian rule) a reversal before reaching it warns that the move in the other direction may travel further than normal. Variants reposition the anchor to handle steep moves: the Schiff pitchfork shifts A halfway toward B in price, and the modified Schiff shifts it to the midpoint of A-B in both price and time. Warning lines are extra parallels spaced beyond the tines to track price that escapes the fork.

The fork's appeal is that it is pure price geometry. There is no lookback length to tune and no smoothing to choose, unlike an [EMA](https://www.luxalgo.com/library/concept/ema/) channel or [MA envelope](https://www.luxalgo.com/library/concept/ma-envelope/); three pivots fully determine every line. That same property is its weakness: pivot selection is subjective, and two chartists can draw materially different forks on the same trend. The discipline that makes it usable is treating a fork that price refuses to respect as wrongly anchored and either re-selecting pivots or standing aside.

## How to draw Andrews' Pitchfork

The fork is only as good as the pivots that anchor it, so start from clean, alternating swing points.

1. Pick pivot A at the origin of the move you want to frame: a significant swing low for an uptrend, or swing high for a downtrend.
2. Pick pivots B and C: after a low at A, B is the next swing high and C the following swing low; after a high at A, B is the next swing low and C the following swing high. A, B and C must alternate.
3. Draw the median line from A through the midpoint of segment B-C, then add parallels through B and C to form the tines.
4. Check the fit: a valid fork should contain the subsequent rotations, with price working between the tines and repeatedly engaging the median line.
5. If price immediately leaves the fork, the anchors are probably wrong: reselect pivots or switch to a Schiff or modified Schiff variant, which flatten the fork for steep moves.

## How it's calculated

A three-pivot channel tool: a median line drawn from an anchor pivot through the midpoint of the next two pivots, framed by parallel tines.

```
1. Pick three alternating pivots in time order: anchor A, then swing extreme B, then C, the swing extreme after B on the opposite side.
2. Find M, the midpoint of the segment from B to C, halfway in both time and price.
3. Median line: a ray drawn from A through M, extended to the right.
4. Tines: rays from B and from C parallel to the median line; the three lines form the fork.
5. Warning lines: optional extra parallels beyond either tine, spaced by the same vertical distance as the tine to the median line.
6. Schiff variant: keep the anchor's time at A but move its price to (price of A + price of B) / 2, then repeat steps 2 to 4.
7. Modified Schiff variant: place the anchor at the midpoint of the segment from A to B, halfway in both time and price, then repeat steps 2 to 4.
8. Inside/outside read: price holding between the tines is inside the fork and keeps the median line as the working reference; sustained closes beyond a tine are outside and argue for acceleration or a redrawn fork.

  A: first pivot, the anchor of the fork
  B: second pivot, the first swing extreme after A
  C: third pivot, the swing extreme after B on the opposite side of price
  M: midpoint of the segment from B to C
```

Andrews' median-line study expects price to return toward the median line after pivot C forms; failing to reach it is itself read as information about trend strength.

The Schiff and modified Schiff variants flatten the fork's slope, which suits steep or accelerating trends.

Placement is entirely pivot-dependent, so different pivot choices produce different forks on the same trend.

## How traders use it

- Trading rotations toward the median line: entries near a tine with the median line as the working objective, on the logic that trends oscillate around their center.
- Reading trend health: repeated failures to reach the median line signal a weakening trend, and a decisive break outside a tine that holds on retest suggests the trend is changing character or accelerating.
- Extending with warning lines when price runs beyond the fork, keeping the original slope as the frame of reference.
- Pairing tine tests with confirmation: a reaction at the lower tine plus an internal [trendline](https://www.luxalgo.com/library/concept/trendline/) break, or agreement from a [moving average crossover](https://www.luxalgo.com/library/concept/moving-average-crossovers/), before committing to the rotation trade.
- Framing risk mechanically: stops go beyond the tine or beyond the anchoring pivot, and targets stage at the median line first and the far tine second, so the fork supplies both invalidation and objectives.

## Andrews' Pitchfork vs other trend frames

- **Trendline** (https://www.luxalgo.com/library/concept/trendline/): A trendline is a single boundary drawn from two pivots and says nothing about where the trend's center or far side sit. The pitchfork commits to a full three-line frame from three pivots, which adds information and adds ways to be wrong.
- **MA Envelope** (https://www.luxalgo.com/library/concept/ma-envelope/): An envelope wraps a moving average at a chosen offset, so its center and width follow parameters and update every bar. The fork's lines are fixed by three historical pivots and never recalculate; only your choice of pivots can change them.
- **Dynamic S/R Via MA** (https://www.luxalgo.com/library/concept/dynamic-s-r-via-ma/): Averages acting as moving support and resistance adapt continuously and carry no directional geometry. The pitchfork fixes a slope from the trend's own structure, so it states where rotations should land if the trend keeps its established pace.

## FAQ

### Which three points do you use for Andrews' Pitchfork?

Three alternating pivots. A sits at the start of the move (a major swing low in an uptrend), B is the next swing high, and C the following swing low; in a downtrend the order mirrors. The median line runs from A through the middle of B-C. Different pivot choices produce very different forks, which is why practitioners redraw with Schiff variants when the standard fork will not contain price.

### Does price always return to the median line?

No. Andrews taught that price tends to return to the median line, and his rules treat the exception as information: when price reverses before reaching it, the move away is often expected to travel further than normal. Treat the median line as a scenario target that needs confirmation, not a certainty, and reassess the fork's pivots when price ignores it entirely.

### What is the Hagopian rule?

It is Andrews' failure rule. When price reverses before reaching the median line it was expected to test, the resulting move in the opposite direction is expected to travel further than a normal rotation. In practice it turns a failed median-line test into a signal in its own right, which is why fork traders pay as much attention to misses as to touches.

### What is the difference between the standard, Schiff, and modified Schiff pitchforks?

They differ only in where the handle pivot A sits. The standard fork uses the raw pivot. The Schiff variant slides A halfway toward B in price, and the modified Schiff slides it to the midpoint of A-B in both price and time. Both adjustments flatten the fork, which helps when the initial move is so steep that the standard median line rises or falls faster than any realistic trend.

### Which timeframes does the pitchfork work on?

The geometry is scale-independent, and practitioners use it from intraday to monthly charts. What changes is pivot quality: higher-timeframe swings are usually less ambiguous, so forks drawn from them tend to be more stable, while intraday forks need frequent redrawing as fresh pivots print. Whatever the timeframe, the fork only means something while price actually respects it.

### How do you trade with the median line?

The common playbook enters at a tine with the median line as the first objective: buy reactions at the lower tine in an up fork, with the stop beyond the tine or the anchoring pivot. Confirmation from momentum or a small structure break improves the odds of catching a genuine rotation. If price stalls repeatedly short of the median line, the Hagopian read warns the trend is weaker than the fork assumes.

## Implementations in the Library

- Andrews' Pitchfork (LuxAlgo): https://www.luxalgo.com/library/indicator/andrews-pitchfork/

## Related concepts

- Trendline: https://www.luxalgo.com/library/concept/trendline/
- Linear-regression Channel: https://www.luxalgo.com/library/concept/linear-regression-channel/
- Parallel Channel: https://www.luxalgo.com/library/concept/parallel-channel/
- Fan Principle: https://www.luxalgo.com/library/concept/fan-principle/
- Standard-error Channel: https://www.luxalgo.com/library/concept/standard-error-channel/
- Polynomial Regression Band: https://www.luxalgo.com/library/concept/polynomial-regression-band/
- Gann Fan & Angles: https://www.luxalgo.com/library/concept/gann-fan-and-angles/
- Gann Box: https://www.luxalgo.com/library/concept/gann-box/
- Gann Square of 9: https://www.luxalgo.com/library/concept/gann-square-of-9/
- Speed Resistance Lines: https://www.luxalgo.com/library/concept/speed-resistance-lines/

---

Source: https://www.luxalgo.com/library/concept/andrews-pitchfork/ (LuxAlgo Library, the encyclopedia of trading & technical analysis). Free to use with attribution: https://www.luxalgo.com/library/license/