Bitcoin risk metric / Power Law Oscillator (AUD)

Historical Risk Metric extremes

The table below shows the months Bitcoin printed its highest and lowest Risk Metric readings. Top readings mark cycle peaks; bottom readings mark cycle bottoms. Both serve as historical anchors for what "cycle top zone" and "cycle bottom zone" actually look like on this metric.

What is Bitcoin's Risk Metric right now?

Bitcoin's Risk Metric for October 2026 is 0.37 (37.5 on the gauge), which the gauge classes as below fair value. For context, the 2017 cycle peaked at 0.89, the 2021 cycle at 0.83 (March 2021) and the 2024-2025 cycle at 0.63 (January 2025). The latest decline bottomed at 0.31 in June 2026. 76 percent of all months have read between 0.3 and 0.7.

What do the Risk Metric zones mean?

The gauge divides the 0-1 scale into 5 classification bands. They are percentile bands of the metric's own monthly AUD history since January 2013 (166 months), recomputed on every data refresh, so the zones describe where a reading sits in Bitcoin's own record rather than borrowing fixed lines from a USD chart:

• 0.77 to 1.0 - Cycle top zone (red): the 95th percentile and above (7 percent of months): November 2013, December 2013, January 2014, February 2014, November 2017, December 2017, January 2018, February 2018, March 2021, April 2021, October 2021.
• 0.64 to 0.77 - Caution zone (orange): 80th to 95th percentile. Late-cycle, above-trend territory.
• 0.43 to 0.64 - Near fair value (amber): 35th to 80th percentile, close to the long-run regression line. The most common reading.
• 0.29 to 0.43 - Below fair value (lime): 10th to 35th percentile. Below-trend territory.
• Below 0.29 - Deep value (green): below the 10th percentile (10 percent of months): January 2013, April 2015, May 2015, June 2015, August 2015, September 2015, October 2015, January 2016, February 2016, March 2016, April 2016, May 2016, July 2016, August 2016, September 2016, October 2016, November 2016.

How the documented cycle tops and bottoms classify under these zones:

What is the Bitcoin risk metric?

The risk metric translates Bitcoin's position on the logarithmic regression chart into a single intuitive 0-1 number. This makes long-term cycle positioning easy to communicate and easy to track over time. Where the log regression chart shows the WHOLE distribution of price-vs-trend, the risk metric collapses it to a single score for the current moment.

The methodology was popularised by Into The Cryptoverse for the USD market. The SatoshiMacro version is the AUD-native equivalent: same conceptual framework, applied to AUD-priced Bitcoin data, recomputed independently on every page load with the methodology fully disclosed.

Three of the five bands carry most of the meaning:

• Below 0.29 (deep value, green zone). Bitcoin trades well below its long-run trend: the lowest 10 percent of monthly readings on the AUD fit.
• 0.43 to 0.64 (fair value, yellow zone). Bitcoin near its regression line. Most of Bitcoin's history has been in this zone or moving through it.
• 0.77 to 1.0 (cycle top, red zone). Bitcoin well above its trend: the highest-scoring months, each within months of a cycle high (November 2013, December 2013, January 2014, February 2014, November 2017, December 2017, January 2018, February 2018, March 2021, April 2021, October 2021).

The Power Law Oscillator (Santostasi)

The same indicator goes by two names depending on the lineage you came in through. In the Into The Cryptoverse / Bitbo / LookIntoBitcoin lineage it's the Risk Metric. In the Giovanni Santostasi lineage it's the Power Law Oscillator (PLO). The mathematics is identical: take the residual of price from a log-log power-law fit (equivalent to a logarithmic regression), normalise to a 0-1 scale, and read cycle position from where the current value sits.

The Power Law framing is now widely adopted because it gives the indicator a mechanistic foundation rather than just an empirical curve fit. The Power Law model proposes that Bitcoin's price follows P(t) ≈ A × t^n (n is about 5.8 in Santostasi's USD fit and 5.41 on this page's AUD data), driven by network-adoption growth (Metcalfe-style) operating against a fixed-supply curve. The Power Law Oscillator measures, at any given moment, how far Bitcoin sits above or below that power-law trajectory, expressed as a 0-1 cycle-positioning score. The two together form the complete Power Law toolkit: the model tells you where the long-run trajectory is; the oscillator tells you where price sits relative to it right now.

Practically the PLO behaves exactly like the Risk Metric you'd read on Into The Cryptoverse for USD. The SatoshiMacro implementation is the AUD-native equivalent, computed independently on every page load from AUD-priced data. On the AUD fit, under the percentile zones, the documented tops read 2013 cycle top 1.00 (Cycle top zone); 2017 cycle top 0.89 (Cycle top zone); 2021 first peak 0.83 (Cycle top zone); 2021 second peak (cycle top) 0.772 (Cycle top zone), and the documented lows read 2015 cycle bottom 0.32 (Below fair value); 2018 cycle bottom 0.34 (Below fair value); 2022 cycle bottom 0.30 (Below fair value).

How does PLO differ from the underlying log regression chart?

• Log regression chart: shows the WHOLE distribution of price-vs-trend across history (every monthly dot plus the central fit and the ±1σ and ±2σ envelopes). Useful for visualising the model and seeing the full shape of cycles. See Bitcoin Power Law / log regression bands chart.
• Power Law Oscillator (this page): collapses the same model to a single 0-1 number for the current moment, suitable for at-a-glance cycle reading and quantitative rules ("trim when PLO > 0.77", "accumulate when PLO < 0.43").

Both views are derived from the same underlying log-log fit. They share strengths (clear cycle-zone identification, three-cycle empirical track record) and limitations (assume future cycles resemble past cycles, monthly resolution only). Pair this indicator with the Mayer Multiple, Pi Cycle Top, and 200-week MA Heatmap for confluence reads, as covered on the Bitcoin & Crypto Charts Dashboard.

Citation. Giovanni Santostasi, "The Bitcoin Power Law Theory", working paper, 2024. Earlier prior art on the underlying log-log fit: Harold Christopher Burger, hcburger.com/blog/powerlaw/ (2019); original Bitcointalk thread by user "Trolololo" (2014).

How to use the metric

The metric is a multi-year-horizon tool. Practical applications:

Long-term accumulation timing. Investors with multi-year holding horizons use the deep-value zone (below 0.29) as the accumulation regime. The framework is "buy when the risk is low" not "buy a specific price". Visits to the deep-value zone have historically lasted months, so the framework gives a window rather than a precise entry point.

Cycle-top profit taking. Investors planning to scale out of positions use the cycle top zone (0.77 and above) as the distribution regime. The framework supports laddered profit-taking (for example, sell 25 percent on entering the caution zone at 0.64 and another 25 percent on entering the cycle top zone at 0.77) rather than calling the top precisely.

Tax-loss harvesting timing for Australian investors. If the risk score is below 0.29 AND you have unrealised losses on Bitcoin, the framework suggests holding rather than realising (because the regime is statistically favourable). If the risk score is at or above 0.77 AND you have unrealised gains, the framework suggests considering CGT-discount-aware profit-taking. The CGT Calculator and Tax-Loss Harvesting Calculator on SatoshiMacro handle the after-tax math.

Portfolio rebalancing trigger. Some long-term investors use the metric as a rebalancing trigger: trim Bitcoin exposure when risk crosses 0.77, restore exposure when risk falls below 0.29. The framework provides a disciplined rule rather than emotional timing.

Methodology

1. Take BTC/AUD monthly closes from January 2013 (the last daily close of each month: Bitstamp BTC/USD converted at the daily AUD/USD rate, with Kraken's native BTC/AUD market for the last two years).
2. Fit a logarithmic regression: log10(price) = slope × log10(days since Bitcoin genesis) + intercept. Use ordinary least squares.
3. Compute the residual for each historical month: residual = actual log10(price) - predicted log10(price).
4. Compute the residual standard deviation σ across all months.
5. For each month (including the current), compute the sigma-deviation: deviation = residual / σ.
6. Clamp the sigma-deviation to [-3, +3] (anything beyond is treated as the extreme).
7. Linearly map the clamped value from [-3, +3] to [0, 1]: risk = 0.5 + deviation / 6.
8. Classify each month into five zones using the 95th, 80th, 35th and 10th percentiles of the full monthly series as cut-offs (currently 0.77, 0.64, 0.43 and 0.29). The cut-offs are recomputed on every data refresh and shared with the embed widget and preview card.

The result is a 0-1 score that updates monthly. The full historical series of risk scores is plotted in the historical chart above; the current score is displayed in the gauge.

The regression is re-fitted on every page load, which means the slope, intercept, and σ all incorporate the latest available data. This is conservative (the model uses the most recent data even if it has only just become available); some implementations use a fixed-from-date regression which can drift if the slope changes structurally over time.

Where the metric breaks down

• It is a one-factor model. Real-world Bitcoin cycle dynamics involve macro liquidity, regulation, halvings, ETF flows, and more. The risk metric collapses all of this into a single number derived from price-vs-time. Practitioners often supplement it with macro overlays.
• The bands are statistical, not causal. Visiting the cycle top zone does not cause a cycle top; it is statistically associated with cycle tops in the historical sample of three full cycles. Because the zones are percentiles of that sample, they will shift as more cycles are added.
• Refitting the regression as each new month arrives means the historical risk series changes slightly over time. Today's risk score for a month in 2017 is not exactly the same as it was when computed in 2017 (because the regression has been refitted with more data since). This is intentional and conservative but worth understanding.
• The metric is monthly resolution. Within a month, the score does not change. For intraday or weekly trading, the metric is too coarse. It is designed for multi-year investors.
• AUD-USD FX effects mean the AUD version diverges from the USD version when AUD is at the extremes of its long-term range. Both can be correct for their respective audiences; the AUD version is correct for AUD-resident investors who measure portfolio value in AUD.

Related tools

• Bitcoin Logarithmic Regression Bands (AUD) - the underlying chart this metric is derived from. Shows the full distribution of historical price-vs-trend rather than the collapsed 0-1 score.
• Bitcoin Rainbow Chart (AUD) - sentiment-labelled version of the same regression. Risk metric value maps roughly to rainbow band classification.
• Bitcoin Pi Cycle Top Indicator (AUD) - top-zone confirmation when the risk metric reads 0.77 or more.
• Bitcoin Mayer Multiple (AUD) - price ÷ 200DMA. Cross-reference with the risk metric for cycle phase.
• Bitcoin Dominance Chart - rotation context. Risk metric peaks have historically coincided with dominance peaks.
• Altcoin Season Index - flow companion to dominance for the BTC-vs-alts rotation read.
• Bitcoin Monthly Returns Heatmap (AUD) - month-by-month percentage returns colour-coded by magnitude. Complementary to the risk metric for understanding cycle volatility.
• Crypto CGT Calculator - applies the ATO 50 percent discount to a specific Bitcoin disposal. Use after risk-metric-informed exit timing.
• Tax-Loss Harvesting Calculator - estimate EOFY tax savings; relevant when the risk metric suggests holding through a drawdown.

Frequently asked questions

About the author

Govind Satoshi

Former Institutional Trader. Founder, SatoshiMacro.

Traded allocated institutional capital at a Sydney proprietary trading firm.