The Ergodic Indicator (TSI): What It Is and Whether It Actually Works

What the Ergodic Indicator (TSI) Actually Is

The True Strength Index (TSI) was developed by William Blau and published in the early 1990s. You'll also see it called the Ergodic Oscillator or Ergodic TSI — the names overlap because Blau built a family of double-smoothed momentum indicators under the ergodic label, and the TSI is the most widely implemented member of that family.

The core mechanism: TSI takes a period-over-period price change, smooths it twice with exponential moving averages, then divides by the same double-smoothed version of the absolute price change. That ratio — scaled to roughly −100 to +100 — gives you a bounded oscillator that moves more slowly than raw RSI or standard momentum lines. Zero-line crosses serve as trend signals. Divergence between price and TSI is the other main setup.

The appeal is obvious on a chart. The line is clean, signals are infrequent, and it looks like noise has been filtered out. Whether that visual calm reflects genuine edge is a separate question — and it's one almost nobody publishing about the TSI actually answers.

Why You Can't Find an Honest Review

Search for 'ergodic indicator' and you get formula reposts. Pages explain how to calculate it, how to interpret the zero line, which default parameters to use. Almost none of them show systematic performance data across a real range of assets and timeframes. That gap is why searchers end up frustrated.

The assumption embedded in most TSI coverage is that double smoothing is better than single smoothing by construction — that less noise means more edge. That logic sounds right until you account for what double smoothing actually costs: lag. By the time the TSI registers a meaningful directional move, a material portion of that move may already be behind you. Whether the remaining portion justifies a trade — after spreads and commissions — is exactly what a structured backtest is designed to test.

Visual smoothness and actual edge are not the same thing. This is one of the most reliable ways technical indicators mislead traders.

What 660,005 Backtests Actually Show

We ran 660,005 out-of-sample backtests across 903 assets — stocks, forex, crypto, commodities, ETFs, and indices — testing 382 indicators on the 1-Hour, 4-Hour, Daily, and Weekly timeframes, with realistic transaction costs included throughout.

The TSI-family did not surface as a top performer in any of our seven asset-class rankings. Across all 903 assets and all indicators tested, only 26% of indicator-asset combinations beat buy-and-hold on a risk-adjusted basis. The median best Sharpe ratio among strategies that did beat buy-and-hold was 0.62 — useful context for how high the bar actually is.

What topped those rankings? In forex, the Fisher Transform led on 17 assets. In stocks, Fibonacci Pivots topped 22 assets. In crypto, MA Envelope led on 5 assets. These results came from empirical testing, not from intuition about which indicators look smoother on a chart. You can see the full breakdown by asset at our results page.

The High Win-Rate Trap That Catches Oscillators

Here's the pattern you'll see repeatedly once you look at indicator data systematically: smoothed oscillators tend to post high individual-trade win rates while still failing to beat passive buy-and-hold. The two metrics are not the same thing, and confusing them is expensive.

Our backtest data illustrates this clearly. The Money Flow Index posted a median win rate of 72.2% across the assets where we tested it — but only 9% of those assets actually beat buy-and-hold. The CCI: 71.0% wins, 9% beat-rate. The Ultimate Oscillator: 72.7% wins, 11% beat-rate. High win rates with low beat-rates is the signature of an indicator that cuts losses quickly and small while also cutting winners early — the net result is lag that neutralizes the apparent signal.

The TSI's double smoothing is doing the same thing mechanically. It reduces the number of losing trades by being slower to enter. But it also reduces the size of winning trades for the same reason. Whether that tradeoff works in your favor depends on the specific asset and market regime — and most of the time, in most regimes, the data says it doesn't clear the buy-and-hold bar.

The Honest Verdict

The Ergodic/TSI is a coherent piece of math. Double-smoothed momentum normalization is a principled idea. But 'principled' doesn't mean 'effective' across a real portfolio, and the performance record in our data reflects that gap.

If you're drawn to the TSI because you want less noise in your momentum signal, the practical question is: less noise compared to what, and at what cost in lag? The indicators that empirically outperformed in each asset class — findable here — tended to win not by being smoother but by being better calibrated to how that specific market actually moves.

Check what actually works for the asset you're trading. The answer is more specific than any generic oscillator can be, and it's grounded in out-of-sample data rather than visual appeal. Browse by asset class here to see what led each category in our tests.

Important: All results described here are hypothetical backtest results run on historical data with transaction costs included. They are not real trading performance and do not guarantee future returns. Nothing on this site is financial advice. All trading involves risk of loss.