Parabolic SAR
A trailing stop that accelerates. It is always in the market, always on one side, and it flips when price reaches it.
MadStockAlerts Research · Updated August 28, 2026
What to take away
- It plots a series of dots that trail price and accelerate toward it.
- When price touches the dots, the indicator flips to the other side.
- It is always long or short, with no neutral state.
- The acceleration factor is the parameter, and it is what causes whipsaws.
- It is a stop mechanism rather than an entry signal.
MAD Academy Training Video · 0:44
A Stop That Only Moves One Way
Parabolic SAR is a trailing stop with an accelerator built in, and it is designed to eventually be hit.
This lesson is part of a Stock Alerts + Tools plan.
What it does
Stop and reverse is the name and the description. The indicator maintains a level on one side of price, moves that level toward price each period, and reverses to the other side when price reaches it.
The rate at which the level advances is governed by an acceleration factor that increases each time a new extreme is made, which is why the dots converge on price faster the longer a move persists.
- 1A trend beginsThe level trails behind price
- 2Each new extreme accelerates itThe gap between price and the level narrows
- 3Price touches the levelWhich is the stop being hit
- 4It flips to the other sideImmediately, and a new position is implied
Why it whipsaws
The acceleration is the mechanism and the weakness. In a sustained trend, tightening toward price locks in more of the move; in a range, it guarantees the level is reached repeatedly, and each touch is a reversal.
Because the indicator has no neutral state, a range produces a continuous sequence of reversals, each one a full position change. That is the most expensive possible behaviour in the market condition that is most common.
Using it as a stop
The construction is far more defensible as an exit than as an entry. A level that trails a move and tightens as the move extends is a reasonable description of a trailing stop, and the acceleration reflects the sensible idea that a move that has run further has more to give back.
- Used as a trailing stop, its always-in property is irrelevant, because the flip is simply the exit.
- It needs a trend condition alongside it, which is what an ADX filter is conventionally used for.
- The acceleration factor and its maximum are the two parameters, and optimising them across history is fitting.
- It is computed from highs and lows, so it responds to gaps.
What it is not
The flip is not a forecast that the direction has changed. It is a statement that price reached a level that was trailing it, which is exactly what a stop being hit means and carries the same information: the previous condition no longer holds.
The parameters, and what changing them does
Two numbers govern the indicator: the step by which the acceleration factor increases at each new extreme, and the maximum it can reach.
| Setting | Effect |
|---|---|
| A larger step | The level converges on price faster, so exits are earlier and whipsaws more frequent |
| A smaller step | The level trails further back, so more of a move is given back before exiting |
| A higher maximum | The level can eventually converge very close to price in a long trend |
| A lower maximum | The level stops tightening, leaving more room in an extended move |
Every row is the same trade-off between capturing more of a trend and giving back less at the end of one, expressed through a different parameter. There is no setting that avoids it.
Because the indicator is used as an exit, the choice is equivalent to choosing how wide a trailing stop should be, which is a risk decision rather than an indicator setting. Framing it that way makes the trade-off explicit.