How is the Premier Stochastic different from a regular stochastic oscillator?

It is a regular stochastic underneath, but the %K is centered, double-smoothed with EMAs, and compressed through a sigmoid so it runs between -1 and +1. The practical differences are fewer mid-range wiggles, sharper transitions between extremes, and a bar or two of extra lag from the smoothing. The information source, where price closed within its recent range, is unchanged.

What do the ±0.2 and ±0.9 levels mean on the Premier Stochastic?

They are convention rather than magic numbers. The ±0.9 lines mark the stretched zone where momentum sits near its recent extreme, and ±0.2 bounds the neutral band whose crossings flag a directional shift. In trending markets the line can hold beyond ±0.9 for long stretches, so exits from that zone, rather than entries into it, are the more common trigger.

Who created the Premier Stochastic?

Lee Leibfarth, a trader and systems developer, published it in the August 2008 issue of Technical Analysis of Stocks & Commodities. The design belongs to that era's transform wave, alongside Ehlers' Fisher family, applying signal-processing squashes to familiar oscillators, and it spread through platform ports that kept his stages and threshold conventions largely intact.

What exactly does the final transform stage do?

It maps the smoothed, centered stochastic through an exponential ratio of the tanh family: values near zero pass through almost linearly while large values compress asymptotically toward ±1. That is what manufactures the indicator's personality, brisk travel through the middle, flattening saturation at the extremes, and it is the same mathematical move the inverse Fisher transform applies to oscillators generally.

Why does the line flatten near its extremes?

Sigmoid saturation: past a point, ever-larger smoothed inputs produce ever-smaller output increments, so once momentum is strongly one-sided the line parks near ±1 and stays there while conditions persist. The flattening is informative twice, arrival at the zone marking stretched momentum, and the eventual crossing back out marking its fade, which is why the exit from the extreme is the conventional event.

What is the Premier RSI?

The same pipeline with RSI as the input: center the oscillator, double-smooth with EMAs, compress through the sigmoid to ±1, mark the same threshold pairs. The port, popularized by LazyBear's implementation, demonstrates the design's general claim, that the center-smooth-squash recipe upgrades the manners of any bounded oscillator whose raw form wobbles too much for clean threshold rules.

Concept

Premier Stochastic

Premier Stochastic is a Momentum & Oscillators concept. The Library holds 1 implementation, a working definition you can pull into Quant.

Top Premier Stochastic indicator

The top custom implementation, built on the original standard Premier Stochastic formula.

1 total

Want to trade Premier Stochastic? The implementation below is one prompt away from a backtested strategy in Quant.

Premier Stochastic

Indicator

What is the Premier Stochastic?

The Premier Stochastic is a smoothed, rescaled variant of the stochastic oscillator, introduced by Lee Leibfarth in a 2008 Technical Analysis of Stocks & Commodities article. It takes a short-lookback %K (eight bars in the original), centers and rescales it around zero, double-smooths it with exponential moving averages, then compresses the result through an exponential normalization from the same sigmoid family as the inverse Fisher transform. The output runs between -1 and +1.

The transform is the point. A raw stochastic wobbles through the middle of its range, which makes threshold rules noisy; the sigmoid steepens movement through the center and flattens it near the bounds, so the line tends to travel decisively between extremes and then level off. Standard plots mark inner thresholds at ±0.2 and outer thresholds at ±0.9, read as momentum-shift and stretched-momentum zones. The smoothing that cleans the line also adds lag, the usual trade.

The pipeline rewards a step-through. The eight-bar %K locates the close within its recent range; subtracting 50 and scaling by a tenth centers that reading around zero in manageable units; two successive EMA passes smooth the centered series, the double smoothing being where most of the noise dies and most of the lag is born; and the final stage pushes the smoothed value through an exponential ratio, the tanh-family squash, that maps everything into the open interval between -1 and +1. Each stage is generic; the specific chaining is Leibfarth's recipe, and the personality, decisive travel with flat extremes, falls out of the sigmoid's saturation.

The design generalizes beyond its stochastic core, which is the second thing worth knowing: the same center-smooth-squash pipeline applies to any bounded oscillator, and the Premier RSI port runs it on RSI input to the identical effect. In deployment the Premier behaves as a cleaner-mannered stochastic, its saturation zones riding trends the way pinned oscillators do, so the exits from extremes outrank the entries into them, and a regime filter in front decides whether threshold crossings are reversion signals or trend noise.

How to identify the Premier Stochastic

Four pipeline stages and two threshold pairs: the construction is a recipe, and the recipe is checkable.

1Compute the short stochastic: an eight-bar %K locating the close within its recent range.
2Center and scale: subtract 50 and multiply by 0.1, moving the series to a zero-centered working range.
3Double-smooth with EMAs, the stage that trades noise for lag.
4Compress through the exponential sigmoid, mapping the smoothed series into -1 to +1 with saturating extremes.
5Mark the conventional thresholds, ±0.2 inner and ±0.9 outer, and read crossings of them rather than the line's wiggles.

How it's calculated

A double-smoothed stochastic passed through an exponential normalizer so the output oscillates between -1 and +1.

\%K_t = 100 \times \frac{C_t - \operatorname{LL}_n}{\operatorname{HH}_n - \operatorname{LL}_n}

\operatorname{NK}_t = 0.1 \times (\%K_t - 50)

\operatorname{SK}_t = \operatorname{EMA}(\operatorname{EMA}(\operatorname{NK}, \text{len}), \text{len})

\operatorname{PSO}_t = \frac{\exp(\operatorname{SK}_t) - 1}{\exp(\operatorname{SK}_t) + 1}

C_t: closing price of bar t

t: bar index

n: stochastic lookback length (default 8)

LL_n: lowest low of the last n bars

HH_n: highest high of the last n bars

%K_t: raw stochastic value at bar t

NK_t: normalized stochastic centered on 0; NK denotes the series

m: smoothing period (default 25)

len: EMA length, round(sqrt(m)) (5 when m = 25)

EMA(x, len): exponential moving average of series x over len bars

SK_t: double-smoothed normalized stochastic

exp(x): exponential function e^x

PSO_t: Premier Stochastic value at bar t, bounded between -1 and +1

Published by Lee Leibfarth in Technical Analysis of Stocks & Commodities, August 2008.

The final transform equals tanh(SK_t / 2), which is what caps the output inside -1 to +1.

Common signal thresholds are +/-0.2 and +/-0.9.

How traders use it

Threshold signals: crossings back out of the ±0.9 extreme zones are read as momentum fading from a stretched state, and crossings through the ±0.2 band as fresh directional shifts. Neither is a standalone entry; in a strong trend the line can ride an extreme for many bars.
As a drop-in replacement for a raw stochastic in overbought/oversold logic when the raw line flips too often: the double EMA smoothing suppresses one-bar whipsaws at the cost of slightly later signals.
Divergence reads: as with other bounded oscillators, a new price high the Premier Stochastic fails to match is read as thinning momentum, with the usual caveat that divergences can persist before they matter.
As a transform recipe: the same center-smooth-squash pipeline applies to other oscillators, the Premier RSI being the established port, so the design travels wherever a noisy bounded series needs decisive extremes.
Regime-gated: threshold crossings are honored as reversion signals only when a trend filter reads conditions as rotational, since the saturating extremes ride trends exactly the way pinned oscillators do.

Premier Stochastic vs related oscillators

Stochastic Oscillator: The raw ingredient: same range-position information, delivered with mid-range wobble and twitchy threshold behavior. The Premier's pipeline trades a bar or two of lag for decisive travel and stable extremes, changing the manners rather than the message.

Inverse Fisher Transform: The compression stage as a standalone concept: the IFT squashes any oscillator toward binary extremes. The Premier is one packaged application, a specific stochastic, specific smoothing, specific thresholds, of the same sigmoid idea.

Stochastic RSI: Opposite treatments of oscillator noise: StochRSI stretches sensitivity, amplifying every wiggle into range-spanning swings, while the Premier suppresses wiggle and saturates extremes. One buys earliness with chatter, the other buys clarity with lag.

Concept family

Momentum & Oscillators

91 concepts mapped · 91 in the Library

Premier Stochastic FAQ

Turn Premier Stochastic into a trading strategy.

Take the implementation from this page into Quant, then build on it, backtest it on real data, and keep refining it in conversation.

Previous conceptPPO Next conceptPretty Good Oscillator