Parabolic SAR on S&P 500: 9 Parameter Combinations Tested

J. Welles Wilder introduced the Parabolic SAR in 1978 as a stop-and-reverse system designed to keep traders on the right side of a trend. The Acceleration Factor (AF) controls how aggressively the trailing stop chases price — Wilder’s original recommendation was AF start = 0.02, max = 0.20. Our 25-year backtest on the S&P 500 (January 2000 – December 2025) tested nine AF combinations to determine whether any configuration beats a passive index fund. The verdict: the best performer, AF 0.01/0.10, achieved CAGR 4.30% and Sharpe 0.420 — below buy-and-hold [2] on both return and risk-adjusted metrics.

Here’s the thing most people miss about this topic.

Wilder designed SAR for trending commodity markets. This test asks what happens when you apply it to the world’s most widely watched equity index over a quarter century.

What Is the Parabolic SAR?

Parabolic SAR plots a trailing stop point above or below price that accelerates toward price over time. When price crosses the SAR level, the system reverses position — from long to short (or in a long-only variant, from long to cash).

Related: cognitive biases guide

The two key parameters:

• AF Start: The initial acceleration factor applied when a new trend begins. Wilder’s default: 0.02.
• AF Max: The ceiling the acceleration factor reaches after successive higher highs (in an uptrend) or lower lows (in a downtrend). Wilder’s default: 0.20.

Each time price makes a new extreme in the trend direction, AF increases by its starting increment — accelerating the SAR toward price. The result is a stop that starts loose and tightens progressively, locking in gains as a trend matures. In a long-only implementation: buy when price crosses above SAR; exit to cash when price crosses below SAR.

The logic is elegant: small AF gives the trend room to breathe early, while the acceleration mechanism ensures you do not give back all profits if the trend reverses late in its cycle.

Backtest Setup

The test period spans four major drawdown events: dot-com crash (2000–2002), financial crisis (2008–2009), COVID crash (March 2020), and the 2022 rate-hike selloff. A trend-following system’s ability to exit before or during these events is the key differentiator.

Buy-and-hold benchmark: CAGR 6.17%, Max Drawdown −56.78%, Sharpe 0.407, final value $4,738,968.

Full Results: All 9 AF Combinations

Green row = best SAR configuration. Negative CAGR indicates capital loss over the full period. All results: 2000–2025, $1M initial capital, no transaction costs.

The Best Configuration: AF 0.01 / Max 0.10

Among nine combinations, AF start = 0.01 with AF max = 0.10 is the clear winner — and the only configuration that achieves a Sharpe Ratio (0.420) marginally above buy-and-hold (0.407). Yet the return gap tells a stark story:

The AF 0.01/0.10 configuration wins on Sharpe by a razor-thin margin (0.420 vs 0.407) and meaningfully reduces the worst drawdown (−43.80% vs −56.78%). However, the CAGR deficit is severe: 1.87 percentage points per year. Over 25 years on a $1M portfolio, that gap compounds to a $1,748,454 shortfall versus buy-and-hold.

The low AF start (0.01) keeps the SAR loose early in a trend — giving positions room to develop without premature exits. The low AF max (0.10) prevents the stop from tightening too aggressively in mature trends, reducing whipsaw. Together they produce the system’s lowest trade count among all configurations and the best win rate at 52.6%.

Wilder’s Original Parameters vs. the Data

Wilder’s default settings — AF start = 0.02, AF max = 0.20 — perform surprisingly poorly on the S&P 500. The AF 0.02/0.20 combination produces negative CAGR (−0.09%), 329 trades over 25 years, and a max drawdown of −55.22% — barely better than buy-and-hold’s −56.78% drawdown while sacrificing all of its return.

This is not a failure of Wilder’s design. When Wilder developed SAR in the 1970s, he was trading commodity futures — markets with genuine trending characteristics, use, and no long-term upward bias. The S&P 500 is structurally different: a capital-weighted basket of U.S. large-cap equities with a persistent long-term upward drift driven by corporate earnings growth and reinvested dividends.

Wilder’s intent was a stop-and-reverse system for futures, not a market-timing tool for equity indices. Applied to equity index trend-following, his original parameters generate excessive trading frequency (329 signals in 25 years = 13+ per year) that erodes returns through whipsaw without sufficient trend capture to compensate.

The AF Parameter Effect: A Systematic Pattern

The results reveal a clear pattern across all nine combinations:

• Lower AF start dramatically improves performance. Every AF 0.01 combination outperforms its AF 0.02 and AF 0.03 equivalents on CAGR and Sharpe. The difference between AF start = 0.01/max = 0.10 (CAGR 4.30%, Sharpe 0.420) and AF start = 0.02/max = 0.10 (CAGR 0.58%, Sharpe 0.109) is enormous — 3.72 percentage points of annual return from changing a single parameter.
• Lower AF max beats higher AF max within each AF start tier. For AF start = 0.01, the max = 0.10 variant (CAGR 4.30%) leads max = 0.20 (CAGR 3.12%) and max = 0.30 (CAGR 3.14%). Lower AF max prevents the trailing stop from tightening too close to price in mature trends — a critical advantage in an index with multi-year bull runs.
• Higher AF start = more trades, worse results. AF start = 0.03 with max = 0.20 generates 402 trades (16 per year) yet delivers only 0.20% CAGR. More signals do not mean better performance here — they mean more friction and more whipsaw losses.
• The AF max = 0.20/0.30 difference is minimal at AF start = 0.01. The nearly identical results for 0.01/0.20 (CAGR 3.12%) vs 0.01/0.30 (CAGR 3.14%) suggest the ceiling matters less when the floor is low — the AF rarely reaches its maximum before a trend reversal triggers an exit anyway.

Why SAR Underperforms on a Long-Running Bull Market Index

Three structural reasons explain why all nine SAR variants trail buy-and-hold on CAGR:

1. Time Out of the Market

A long-only SAR system sits in cash whenever price is below the SAR level. During the S&P 500’s sustained bull markets (2003–2007, 2009–2020, 2020–2021), repeated whipsaw signals force temporary exits that miss recovery rallies. Cash earns nothing while the index compounds. The best SAR variant holds roughly 70–75% of available trading days based on its 192-trade profile — the rest of the time it’s earning zero.

2. Late Entries, Early Exits

SAR entry requires price to close above the SAR level, which lags the actual trend turn. In a fast V-shaped recovery (like March–April 2020), the system may miss the first 5–15% of a rally before triggering a buy signal. Similarly, SAR exits often trigger during normal pullbacks within ongoing bull markets — selling at temporary lows rather than at actual trend reversals.

3. The Upward Drift Problem

Over 25 years, the S&P 500 grew from approximately 1,469 (Jan 2000) to ~5,800 (Dec 2025) despite two crashes exceeding 50%. Any mechanical system that moves to cash during portions of this period must compensate through dramatically better crash avoidance to justify the opportunity cost. SAR reduces max drawdown from −56.78% to −43.80% at best — meaningful protection, but not enough to overcome the CAGR deficit.

Where Parabolic SAR Genuinely Excels

This backtest is specific to mechanical trend-following on the S&P 500 index. Parabolic SAR has well-documented utility in other contexts:

• Dynamic trailing stop management: SAR is widely used not as an entry signal but as a trailing exit mechanism on positions entered via other methods (momentum, breakout). In this role it manages risk without the whipsaw problem of mechanical system entry/exit.
• Trending commodity futures: Wilder’s original application. Oil, gold, and grain futures exhibit sustained directional trends where SAR’s acceleration mechanism performs as designed.
• Individual equities in strong uptrends: A stock exhibiting parabolic price behavior (consecutive higher highs) benefits from SAR’s progressive stop-tightening to lock in profits before trend exhaustion.
• Intraday and swing trading: On shorter timeframes with more trend episodes, SAR generates more actionable signals with better trend-to-whipsaw ratios than on daily index data.

Comparison to Other Strategies Tested

In our S&P 500 strategy series, Parabolic SAR ranks among the lower-performing systematic approaches:

• Dual Momentum: CAGR 9.29%, Sharpe 0.598 — beats buy-and-hold decisively
• Stochastic Oscillator (K=5/25/85): CAGR 7.23%, Sharpe 0.532 — genuine outperformance
• Keltner Channel (ATR×2.0): CAGR 5.84%, Sharpe 0.401 — near-miss, drawdown benefit
• Parabolic SAR (AF 0.01/0.10): CAGR 4.30%, Sharpe 0.420 — drawdown reduction, return sacrifice
• MACD Signal Line: Best CAGR 4.82%, Sharpe 0.341 — underperforms across all variants

Important Caveats

1. No transaction costs. At 192 trades over 25 years (best variant), roughly 7–8 trades per year. On a $1M portfolio, costs are minor. Smaller accounts face proportionally larger friction.
2. No taxes. Active trading generates short-term capital gains. After-tax returns would widen the gap versus buy-and-hold considerably.
3. No slippage modeled. SAR signals trigger on daily closes. Real-world execution on the next open may face gap risk, particularly after volatile sessions.
4. Index only. Results on individual securities, sector ETFs, or other asset classes are likely to differ. SAR was designed for individual commodity contracts, not index composites.
5. Nine combinations is not exhaustive. A wider parameter grid (e.g., AF start = 0.005, finer AF max increments) might identify marginally better configurations — but also increases overfitting risk on historical data.

Practical Takeaway

Among nine parameter combinations, only AF 0.01/0.10 achieves a Sharpe Ratio marginally above buy-and-hold (0.420 vs 0.407). Every other SAR configuration fails on both return and risk-adjusted basis simultaneously. The reduction in maximum drawdown (−43.80% vs −56.78%) is the system’s sole genuine advantage — but it costs 1.87 percentage points of annual return and $1.75M in 25-year wealth accumulation on a $1M initial investment.

Wilder’s Parabolic SAR remains a valuable indicator for trailing stop placement and visual trend identification. As a standalone mechanical entry/exit system on the S&P 500, the evidence from this 25-year test is clear: it does not generate sufficient alpha to justify active management over passive index investing. Use it as a risk management tool, not a return engine.

Last updated: 2026-04-09

References

1. Alberg, D., et al. (2025). Enhancing Trading Performance Through Sentiment Analysis of Financial News Combined with Technical Indicators on S&P 500. arXiv preprint arXiv:2507.09739. Link
2. StockCharts.com (n.d.). Parabolic SAR. ChartSchool – StockCharts.com. Link
3. Wilder, J. W. (1978). New Concepts in Technical Trading Systems. Trend Research. Link
4. AvaTrade (n.d.). Parabolic SAR (Stop And Reverse) Indicator. AvaTrade Education. Link
5. Samco Securities (n.d.). Stop And Reverse (SAR) Strategy: How To Use Parabolic SAR for Trading. Samco Knowledge Center. Link

Related Reading

• How to Open a Brokerage Account
• DCA Strategy for Beginners [2026]
• Basic Car Maintenance Everyone Should Know: Beginner Guide [2026]

Related Posts

• The Dunbar Number: Why You Can Only Maintain 150 Friendships
• Hedonic Adaptation: Why Getting What You Want Never Makes
• Keltner Channel Breakout on S&P 500: ATR Entry Tested