Momentum indicators

Stochastic Momentum Index (SMI): Formula, Settings and a Test Against the Stochastic

William Blau’s SMI measures the close against the middle of the recent range and smooths it twice. The formula, the usual 10, 3, 3 settings, and how it compared with the stochastic oscillator on real data.

· Founder of PatternsWizard
Checked 7 min read
SMI(10, 3, 3) on S&P 500 ETF, dividend-adjusted prices, daily, August 5, 2026 to October 28, 2026. Illustration of how the indicator reads, not a backtest.
On this page
  1. How the SMI is calculated
  2. Settings
  3. SMI vs stochastic oscillator
  4. Does it work?
  5. Frequently asked questions

The Stochastic Momentum Index (SMI) is William Blau’s version of the stochastic oscillator. Where the stochastic measures the close against the low of the range, the SMI measures it against the middle of the range, then smooths both the numerator and the denominator twice with exponential averages. The result is centered on zero, runs from −100 to +100, and moves more slowly than the stochastic. Below: the formula, the settings, and how the SMI and the stochastic compared on the S&P 500 ETF and on Bitcoin over the last 250 bars.

Stochastic Momentum Index (10, 3, 3) on the S&P 500 ETF

SMI(10, 3, 3) on S&P 500 ETF, dividend-adjusted prices, daily, October 30, 2025 to October 28, 2026. Illustration of how the indicator reads, not a backtest.
  • Price, daily candles (hollow up, filled down)
  • SMI(10, 3) (solid line)
  • Signal, EMA(3) of the SMI (solid line, second color)
SMI(10, 3, 3) on S&P 500 ETF, dividend-adjusted prices, daily, October 30, 2025 to October 28, 2026. Illustration of how the indicator reads, not a backtest.

How the SMI is calculated

Take the highest high and lowest low of the last 10 bars. The distance of the close from the midpoint of that range is the numerator; half the range is the denominator. Each is smoothed by an exponential average of 3 bars, then again by a second one of 3 bars, and the ratio is multiplied by 100. A signal line, an exponential average of the SMI, is plotted alongside. This is TradingView’s formula, written with 200 and the full range, which is the same thing.

The same close, read by the stochastic and by the SMI

The same close, read by the stochastic and by the SMI: ten candles between a dashed line at their highest high and one at their lowest low, with a midpoint line halfway. The last close sits four fifths of the way up the range. A first scale, the stochastic %K, runs from 0 at the low to 100 at the high and reads 80. A second scale, the raw SMI, runs from minus 100 at the low through 0 at the midpoint to plus 100 at the high and reads plus 60.
Schematic, not market data.The hypothetical close of the example sits four fifths of the way up its range: the stochastic measures it from the low, the SMI from the middle.

Settings

LuxAlgo notes that platform defaults vary but that 10, 3, 3 is the most common, with ±40 as the usual overbought and oversold marks and a signal line that is typically a 3-period EMA. Blau presented the indicator in “Stochastic Momentum,” in the January 1993 issue of Technical Analysis of Stocks & Commodities (volume 11, pages 11 to 18), as a variant of the double-smoothed stochastic he had described in the same magazine in January 1991, and developed it in his book Momentum, Direction, and Divergence (Wiley, 1995). A MetaTrader version that follows the book uses a 5-bar range and three smoothings of 20, 5 and 3 bars, so settings copied from one platform to another may not match. The levels are read two ways: LuxAlgo calls ±40 overbought and oversold, while TradingView’s help page reads values above +40 as a sign of a bullish trend and below −40 as bearish. Both readings describe a strong move, not a turn.

SMI vs stochastic oscillator

LuxAlgo describes the SMI as a stochastic that “turns later but whipsaws less.” We checked both halves of that sentence on 250 daily bars of the S&P 500 ETF, from October 30, 2025 to October 28, 2026. A crossing is the main line crossing its signal line, counted on each of the 250 bars against the bar before; a quick reversal is a crossing that came three bars or less after the previous one. An entry above or below a level is a close of the main line beyond +40 or −40 (80 or 20 for the stochastic) after a bar inside.

Slow stochastic (14, 3, 3) on the S&P 500 ETF, same window

Stochastic(14, 3, 3) on S&P 500 ETF, dividend-adjusted prices, daily, October 30, 2025 to October 28, 2026. Illustration of how the indicator reads, not a backtest.
  • Price, daily candles (hollow up, filled down)
  • %K, 14 bars, SMA(3) smoothed (solid line)
  • %D, SMA(3) of %K (solid line, second color)
Stochastic(14, 3, 3) on S&P 500 ETF, dividend-adjusted prices, daily, October 30, 2025 to October 28, 2026. Illustration of how the indicator reads, not a backtest.
Indicator and settingsCrossingsQuick reversalsFirst bullish crossing after the March 30 lowFirst bearish crossing after the October 6 highEntries above / below the levels
SMI 10, 3, 3 (±40)46161 bar later6 bars later10 / 8
SMI 14, 3, 3 (±40)46161 bar later6 bars later8 / 6
SMI 20, 5, 5 (±40)2722 bars later7 bars later6 / 3
Fast stochastic 14, 1, 3 (80 / 20)10178Same bar1 bar later20 / 19
Slow stochastic 14, 3, 3 (80 / 20)62331 bar later2 bars later11 / 9

Both halves held on this sample. The default SMI crossed its signal line 46 times against 62 for the slow stochastic, and 16 of those crossings were quick reversals against 33. It was also later at the top: after the October 6, 2026 close of 779.09, its first bearish crossing came six bars later, against two for the slow stochastic. At the March 30, 2026 low of 628.78, the two turned up on the same bar, March 31. The longer 20, 5, 5 setting cut quick reversals to 2, at the cost of one more bar of delay at both turns.

Bitcoin, over its last 250 daily bars, from February 21 to October 28, 2026, gave the same picture: 50 crossings for the default SMI with 21 quick reversals, against 66 and 36 for the slow stochastic. After the September 21 high close at 86,620, the SMI’s first bearish crossing came three bars later and the slow stochastic’s two bars later. On Bitcoin the 14, 3, 3 setting also cut the count, to 38 crossings, which it did not do on the ETF.

Signal-line crossings over 250 daily bars, by indicator and setting

Signal-line crossings over 250 daily bars, by indicator and setting. SMI 10, 3, 3: S&P 500 ETF 46, Bitcoin 50; SMI 14, 3, 3: S&P 500 ETF 46, Bitcoin 38; SMI 20, 5, 5: S&P 500 ETF 27, Bitcoin 25; Fast stochastic 14, 1, 3: S&P 500 ETF 101, Bitcoin 103; Slow stochastic 14, 3, 3: S&P 500 ETF 62, Bitcoin 66. Our computation; the inputs and the method are in the text.
  • S&P 500 ETF
  • Bitcoin
Our computation; the inputs and the method are in the text.S&P 500 ETF from October 30, 2025 and Bitcoin from February 21, 2026, both to October 28, 2026. Fewer crossings mean fewer whipsaws, not better trades.

Does it work?

We have no long-run measurement of the SMI: our bench tests chart patterns detected by rule against random entries, not indicators (see our methodology). The table above shows a trade-off, not an edge: fewer false crossings bought with later signals. Whether that trade pays depends on the market and on what you do after the signal.

Related pages: the stochastic oscillator, the Stochastic RSI, Williams %R, and the list of momentum indicators.

Every pattern's results on one sheet

The only pattern cheat sheet with tested stats: 152 patterns, what the textbook says and what really happened on 6 markets. Free PDF, sent to your inbox.

The PDF by email, plus our newsletter: new tests and results that change. Unsubscribe anytime. Privacy.

Frequently asked questions

What is the Stochastic Momentum Index?

An oscillator by William Blau that measures where the close sits relative to the midpoint of the recent high-low range, double-smoothed with exponential averages. It runs from −100 to +100 and is centered on zero.

What are the best settings for the SMI?

The most common are 10, 3, 3, with ±40 as overbought and oversold levels. Longer smoothing, such as 20, 5, 5, gave far fewer quick reversals on the S&P 500 ETF (2 against 16) but one more bar of delay at each turn. No setting has been shown to work best.

Is the SMI better than the stochastic oscillator?

It is smoother, not better. On the samples on this page it gave fewer crossings and fewer quick reversals than the slow stochastic, but turned down later after the highs.

What does an SMI above 40 mean?

That the close has stayed well above the middle of its recent range. It marks a stretched market, not a reversal: in a trend the SMI can stay above +40 for a long time.

Sources

  1. William Blau, Stochastic Momentum, Technical Analysis of Stocks & Commodities, V.11:1, January 1993
  2. LuxAlgo, Stochastic Momentum Index
  3. TradingView, Stochastic Momentum Index (SMI)
  4. MQL5 Code Base, Stochastic Momentum Index Blau_SMI
  5. StockCharts ChartSchool, Stochastic Oscillator
, Founder of PatternsWizard

Passionate about finance, quantitative trading and data analysis. Amaury built the PatternsWizard test bench, which detects each pattern on full price history and measures what happened next against random entries.

PatternsWizard tests chart and candlestick patterns on full market history, against random entries. How we test patterns.

179 patterns testedRandom-walk self-test passed

Related guides

All momentum indicators articles