Parabolic SAR is a trend-following indicator that plots a calculated trailing level above or below price.
| What you see | What it means |
|---|---|
| Dots below price | SAR is in an upward phase |
| Dots above price | SAR is in a downward phase |
| Dots switch sides | Price crossed the calculated SAR level |
That last line needs care. A switch does not prove that the market has started a new trend. In a range, the dots can change sides several times while price goes nowhere.
I find Parabolic SAR more useful for managing an existing trend than for finding one. It gives me a rule for moving a trailing level. It does not tell me whether the next reversal signal will continue.
What does SAR stand for?
SAR means Stop and Reverse. J. Welles Wilder Jr. introduced it in his 1978 book New Concepts in Technical Trading Systems, the same book that introduced RSI, ATR and ADX. He designed it as part of a system that assumed a position would close at the SAR level and reverse in the other direction.
You do not have to use it that way. The indicator can serve as a trailing reference without turning every flip into a new entry.
Parabolic SAR uses each candle’s high and low. A wick can therefore trigger a flip even if the closing price later returns. If your rules act only after a candle closes, decide that before reviewing the chart.
How Parabolic SAR is calculated
The core formula is short:
Next SAR = Current SAR + AF x (EP - Current SAR)
EP, or Extreme Point, is the highest high reached during the current upward phase or the lowest low reached during the current downward phase.
AF, or Acceleration Factor, controls how quickly SAR moves toward that extreme. With the standard settings, AF starts at 0.02. Each new extreme increases it by 0.02, up to 0.20.
Suppose an upward phase keeps making new highs. EP rises and AF increases. The dots begin to catch price faster. If no new high appears, AF stops increasing, although SAR still moves toward the existing EP.
There are two less visible rules:
- In an upward phase, SAR cannot be placed above the lows of the previous two candles.
- In a downward phase, SAR cannot be placed below the highs of the previous two candles.
When price crosses SAR, the calculation flips. The previous extreme becomes the first dot on the other side, and AF resets to its starting value.
A worked example with real numbers
Here is one step from the EUR/USD chart below. On 18 March 2025, SAR stood at 1.0755. That day’s high of 1.0954 was a new extreme, so EP moved up to 1.0954 and AF rose to 0.18.
Next SAR = 1.0755 + 0.18 x (1.0954 - 1.0755) = 1.0755 + 0.0036 = 1.0791
The chart shows SAR at 1.0791 on 19 March. That level was still below the lows of the two previous candles, so the two-candle rule did not change it.
You do not need to calculate this by hand. The formula matters because it explains why the dots tighten during a persistent move and why a mature trend can be stopped by a relatively small pullback.
How the dots accelerate and flip
One Parabolic SAR cycle
Follow the calculation from the first flip, through the accelerating trail, to the reset above price.
- Start: SAR flipped below price and AF reset to 0.02.
- Trail: new highs raised EP and AF, pulling SAR closer to price.
- Flip: the candle low crossed SAR, which reset above price.
The chart shows daily EUR/USD from 3 February to 28 March 2025. The cycle runs from 14 February to 20 March.
On 14 February, SAR flipped below price and began a new upward phase at 1.0213, the lowest low of the previous downward phase. As EUR/USD made fresh highs, AF increased from 0.02 to 0.18. The dots rose from beneath the move and gradually closed the distance.
On 20 March, the day’s low of 1.0818 dipped just below the projected SAR level of 1.0821, even though the candle closed much higher, at 1.0912. The indicator reset above price at the previous extreme of 1.0954, while AF returned to 0.02. A difference of about two pips was enough to end the phase, which is why the intrabar-versus-close decision matters.
The flip defined a new phase, the dots trailed the move, and the next flip ended it. Nothing in that sequence guarantees a lasting downtrend next.
How I use Parabolic SAR
1. Confirm that the market is actually trending
I read price first. Higher highs and higher lows support an upward trend. Lower highs and lower lows support a downward one. Repeated movement through the same price area suggests a range.
SAR assumes that a trend exists. It does not measure trend strength. ADX can help with that separate question, but the price structure should still make sense without another indicator.
2. Use the dots as a trailing reference
Once price is moving cleanly, SAR advances in the direction of the trend. It rises during an upward phase and falls during a downward one. This can make exits more consistent, but it may also react to a normal pullback.
I decide in advance whether the rule uses:
- Any intrabar touch
- A candle close beyond SAR
- SAR plus a break in price structure
Those rules can produce different exits from the same chart. Switching between them after seeing the outcome makes a test meaningless.
3. Keep the entry separate from the exit
A SAR flip can be part of an entry setup, but I still want a reason for a trend to develop: a breakout, a pullback that holds, or a wider trend already in place. SAR can then manage the exit instead of being asked to identify the regime, time the entry and define the stop.
Why Parabolic SAR fails in a range
The same settings in two market regimes
SAR trails a persistent move. In a range, the same calculation keeps changing sides.
Both charts use EUR/USD daily data and the standard 0.02 / 0.20 settings.
From 20 February to 25 April 2025, the pair rose 9.1% over 45 trading sessions. SAR flipped twice. Most dots stayed below price and trailed the broader move.
From 18 January to 21 March 2023, EUR/USD finished 0.7% lower after another 45 sessions. SAR changed sides eight times. Each flip correctly reported that price had crossed the current level. None could tell whether that crossing would persist.
This is the main weakness of Parabolic SAR. It always chooses a side. A range has no lasting side to choose.
Parabolic SAR settings
The common defaults are:
| Setting | Standard value | What it changes |
|---|---|---|
| AF start or step | 0.02 | Initial sensitivity and the increase after each new extreme |
| AF maximum | 0.20 | The fastest rate at which SAR can approach price |
Platforms may label the first input as acceleration, step, increment or start. Some expose separate start and increment values.
A larger step makes SAR catch price faster. It locks onto short moves sooner but also flips more often. A smaller step gives price more room and reacts later.
Raising the maximum only matters after enough new extremes have pushed AF toward that cap. It may have little effect on a short or hesitant move.
I start with the defaults. I would not change the settings because a faster combination catches one historical turn perfectly. Compare the same rules across trends, ranges and volatile reversals, then count both delayed exits and false flips.
Parabolic SAR with ADX or a moving average
With a moving average, I use direction rather than a crossover recipe:
- Price above a rising average supports an upward bias; SAR below price can trail it.
- Price below a falling average supports a downward bias; SAR above price can trail it.
With ADX, I look for evidence that directional movement is strengthening. A cluster of SAR flips while ADX remains low is usually a reason to distrust the flips, not trade each one.
The extra tool should answer a different question. SAR already shows direction and a trailing level; ADX or price structure can help decide whether that direction is strong enough to follow.
Parabolic SAR vs other trailing stops
Parabolic SAR is one of several ways to trail a position. Each makes a different trade-off:
| Method | How it trails | Strength | Weakness |
|---|---|---|---|
| Parabolic SAR | Moves toward price faster the longer a trend makes new extremes | Tightens automatically in a strong, extended move | Flips often in ranges; can exit on a normal pullback late in a trend |
| Moving average | Follows the average of recent closes | Smooth and easy to read | Reacts slowly after sharp moves; the exit can come well after the turn |
| ATR-based stop | Stays a multiple of average true range away from price | Adapts the distance to current volatility | Does not tighten on its own as a trend matures |
None of them is best in every market. SAR suits persistent trends where you want the stop to catch up with price over time. A volatility-based stop gives a more constant cushion, and a moving average gives a slower, smoother reference.
Using SAR as a stop reference
The plotted dot and an executed stop are not the same thing. Spread, gaps and fast movement can produce a different exit price. The distance between entry and SAR also affects position risk: a distant level creates more exposure at the same position size, while a close one is easier for ordinary noise to hit.
I calculate acceptable loss from the stop distance and position size. I do not move SAR farther away simply because price is approaching it. That would remove the rule at the moment it becomes inconvenient.
Common Parabolic SAR mistakes
Treating every flip as a new trend
A flip only means price crossed SAR. The range example produced eight of them with almost no net movement.
Using SAR to measure trend strength
The indicator shows a phase and a trailing level. It does not measure how strong that phase is.
Changing settings after every losing example
Faster settings solve some late exits by creating more false flips elsewhere. Test the trade-off across many periods.
Ignoring candle highs and lows
SAR responds to intrabar extremes, not only closing prices. This can explain a flip that looks surprising on a close-only line chart.
Assuming the dots predict future price
Every dot comes from existing highs, lows, EP and AF. The indicator reacts and trails. It does not know the next candle.
How to test Parabolic SAR before relying on it
Before trusting any rule based on Parabolic SAR, check how it behaves on your own charts:
- Start with the default
0.02 / 0.20settings. - Pick two periods of history: one clean trend and one sideways range, like the EUR/USD examples above.
- Count the flips in each. Many flips with little net movement show what a range costs.
- Compare two exit rules: an intrabar touch of SAR and a candle close beyond it. Note how often each exits early or late.
- Change one setting at a time, then repeat the same comparison across several markets and periods.
Write the rules down before you look at the results. A rule chosen after seeing the outcome only describes the past.
The rule I keep
Dots below price mean the upward phase is still active. Dots above price mean the downward phase is active. A switch ends the old phase; it does not guarantee the next one will last.
That is why I use Parabolic SAR to trail a trend I can already see, not to manufacture a trend where price has none.
