Overview
This model combines seven independent indicators into a single composite score (0–100) designed to identify Bitcoin cycle bottoms. Rather than relying on any single metric, it uses the principle of confluence — the idea that when multiple independent analytical frameworks align, the resulting signal is substantially more reliable than any individual reading.
The model is calculated entirely from Bitcoin's weekly closing price history. No on-chain data feeds or paid APIs are required. Where on-chain indicators like MVRV or the Puell Multiple are conceptually important, the model uses price-derived proxies that approximate their behavior.
How to Read the Model
| Composite Score | Zone | Interpretation |
|---|---|---|
| 0–20 | Bull Territory | Price is healthy relative to long-term metrics. No bottom signal. |
| 20–40 | Neutral | Some indicators show stress, but no convergence. Correction territory. |
| 40–70 | Approaching Bottom | Multiple indicators entering historically significant zones. Elevated probability of bottom formation. |
| 70–100 | Extreme Fear / Bottom Zone | Strong multi-indicator convergence. Historically, all confirmed cycle bottoms have scored 70+. |
Data: Weekly BTCUSD, June 2012 – February 2026 (742 data points). Last updated: February 27, 2026.
Core Thesis
Bitcoin's price follows a roughly four-year cycle tied to the halving schedule, which cuts new BTC issuance by 50% approximately every 210,000 blocks. Each cycle follows a pattern:
- Halving reduces supply issuance
- Expansion phase drives price to a new all-time high (typically 12–18 months post-halving)
- Distribution phase as early investors take profits
- Contraction phase as overleveraged positions unwind, weak hands capitulate, and price corrects 70–85% from the peak
- Accumulation phase as long-term holders absorb supply at depressed prices
The average time between halvings is 1,388 days (~3 years, 9 months). The average duration from halving to cycle top is 480 days (~16 months). The average duration from cycle top to cycle bottom is 383 days (~12.5 months). The average duration from cycle bottom to next halving is 523 days (~17 months).
Source: Bitcoin Suisse halving cycle analysis
This model aims to identify phase (4) → (5) transitions — the moment of maximum pain that precedes the next accumulation phase.
The Bitcoin Cycle Framework
Historical Cycle Bottoms
| Cycle | All-Time High | ATH Date | Cycle Bottom | Bottom Date | Drawdown | Days ATH→Bottom |
|---|---|---|---|---|---|---|
| 1 | $31.91 | Jun 2011 | ~$2.01 | Nov 2011 | -93.7% | ~155 days |
| 2 | $1,177 | Nov 2013 | ~$175 | Jan 2015 | -85.1% | ~407 days |
| 3 | $19,783 | Dec 2017 | ~$3,195 | Dec 2018 | -83.8% | ~363 days |
| 4 | $69,045 | Nov 2021 | ~$15,500 | Nov 2022 | -77.6% | ~376 days |
| 5 | $126,296 | Oct 2025 | ? | ? | -46.2%* | ongoing |
*As of Feb 27, 2026. Current price ~$68,000.
Each successive cycle produces a smaller maximum drawdown (93% → 85% → 84% → 78%), reflecting Bitcoin's maturation and deeper liquidity. This "diminishing volatility" trend is critical context for interpreting indicators calibrated to historical extremes.
Sources: Newhedge Bitcoin Drawdown chart, Galaxy Digital research (Feb 2026)
Indicator 1: Price / 200-Week Moving Average
Theory
The 200-week moving average (200 WMA) is the most widely referenced long-term support level in Bitcoin analysis. It smooths out four years of weekly closing prices — almost exactly one full halving cycle — producing a slowly rising line that represents Bitcoin's long-term trend.
The indicator calculates a simple ratio:
When this ratio falls below 1.0, price is trading below the 200 WMA, which has historically coincided with cycle bottoms. When it falls below 0.7, Bitcoin is deeply undervalued by this measure.
Historical Evidence
| Cycle Bottom | 200 WMA at Bottom | Price at Bottom | Ratio | Held as Support? |
|---|---|---|---|---|
| Jan 2015 | ~$289 | ~$175 | 0.61 | Briefly pierced |
| Dec 2018 | ~$3,410 | ~$3,195 | 0.94 | Touched |
| Mar 2020 (COVID) | ~$6,200 | ~$3,850 | 0.62 | Briefly pierced |
| Nov 2022 | ~$22,600 | ~$15,500 | 0.69 | Pierced for ~30 weeks |
In the current cycle, the 200 WMA sits near ~$58,000 (as of Feb 2026), and Bitcoin recently closed a weekly candle below its 200-week EMA for the first time since October 2023, ending an 882-day support streak.
Sources: Nasdaq/Bitcoin Magazine (Jun 2025), TradingView/Cointelegraph (Feb 2026)
Scoring
| Condition | Points |
|---|---|
| Ratio < 1.0 (below 200 WMA) | +15 |
| Ratio < 0.7 (deep undervaluation) | +15 additional |
Implementation
The period is adjustable via the controls panel (150–250 weeks).
Indicator 2: Pi Cycle Bottom
Theory
The Pi Cycle Top Indicator was created by Philip Swift (founder of Bitcoin Magazine Pro) in April 2019. It uses the 111-day moving average and the 350-day moving average multiplied by 2. When the 111DMA crosses above the 350DMA×2, it has historically signaled the exact cycle top within 3 days.
The model adapts this concept for bottoms using weekly equivalents:
- 21-week MA (≈ 147 days, approximating the 111DMA)
- 67-week MA (≈ 469 days, approximating the 350DMA)
When the short-term MA (21w) crosses below the long-term MA (67w), momentum has collapsed to a degree historically associated with bottoming phases. The cross itself does not pinpoint the exact bottom, but the period during which the 21w MA remains below the 67w MA encompasses the bottom formation zone.
The name "Pi Cycle" comes from the ratio 350/111 ≈ 3.153, which is remarkably close to π (3.14159).
Historical Evidence
The Pi Cycle Top Indicator has correctly identified every major Bitcoin cycle top within a 3-day window:
- April 2013: Top at $259 — signaled within 3 days
- December 2013: Top at $1,177 — signaled within 3 days
- December 2017: Top at $19,783 — signaled within 3 days
- April 2021: Top at $64,863 — signaled within 1 day
- October 2025: Top at ~$126,296 — signaled within ~3 days
However, analysts like Benjamin Cowen have noted that diminishing returns are compressing the distance between the Pi Cycle moving averages, causing the bottom indicator variant to trigger earlier than actual bottoms in recent cycles. In 2021, the bottom signal fired approximately 5 months before the actual low.
Sources: Nasdaq/Bitcoin Magazine Pro (Nov 2024), Bitcoin Magazine Pro Pi Cycle chart
Scoring
| Condition | Points |
|---|---|
| 21w MA is below 67w MA (cross active) | +15 |
Implementation
Indicator 3: MVRV Z-Score Proxy
Theory
The true MVRV Z-Score is an on-chain metric created by Murad Mahmudov and David Puell, comparing Bitcoin's Market Value (market cap) to its Realized Value (the sum of the price at which each coin last moved on-chain). The Z-Score standardizes this ratio to identify statistical extremes.
Since on-chain data is not available in this model, we approximate the MVRV Z-Score using price data:
The 4-year moving average serves as a proxy for realized value because it represents the average price across one complete halving cycle, capturing the aggregate cost basis of market participants over that period.
Historical Evidence
The true MVRV Z-Score has identified every major cycle top and bottom:
- Z > 7: Historically signals cycle tops (overvalued)
- Z < 0: Historically signals undervaluation (bottom zone)
- Z < -0.5: Deep undervaluation — has coincided with exact cycle lows
As of February 2026, the actual MVRV Z-Score stands at approximately -2.28, having entered the "strong bear" zone below the levels seen at the 2018 and 2022 cycle bottoms. This deep reading reflects the increased cost basis caused by institutional capital inflows (ETF purchases since 2024).
Sources: Bitcoin Magazine Pro MVRV Z-Score, Glassnode Studio, Bitcoin Magazine (Jun 2025), AInvest (Feb 2026)
Scoring
| Condition | Points |
|---|---|
| Z-Score proxy < 0 (undervalued) | +10 |
| Z-Score proxy < -0.5 (deeply undervalued) | +10 additional |
Implementation
Indicator 4: Logarithmic Regression Band
Theory
Bitcoin's price follows a logarithmic growth curve when plotted over time. This is expected for a technology in its adoption S-curve phase — rapid early growth that gradually decelerates as the market matures.
A logarithmic regression fits the equation:
Where a is the intercept and b is the slope, both fitted via ordinary least squares across all available data points. The lower regression band (mean - 1.5 standard deviations of residuals) defines the statistical floor of Bitcoin's price trajectory.
When price enters the lower band, it is trading at an extreme discount to its long-term growth trend.
Historical Evidence
The logarithmic regression model was first developed by Bitcointalk user "trolololo" in 2014 and later refined by analysts including Benjamin Cowen and CryptoRiskAlert. The lower band has captured every major bear market low:
- 2015 bottom: Price entered the lower band
- 2018 bottom: Price touched the lower band
- 2020 COVID crash: Briefly pierced the lower band
- 2022 bottom: Price entered the lower band
One important caveat: the regression must be periodically re-fitted as new data becomes available. The bands shift slightly with each additional cycle of data.
Sources: TradingView CryptoRiskAlert indicator, Benjamin Cowen logarithmic regression analysis
Scoring
| Condition | Points |
|---|---|
| Price is at or below the lower regression band | +15 |
Implementation
The band width parameter (default 1.5σ) is adjustable via the controls panel (1.0–2.0σ).
Indicator 5: Weekly RSI
Theory
The Relative Strength Index (RSI), developed by J. Welles Wilder in 1978, measures the speed and magnitude of recent price changes to evaluate overbought or oversold conditions. It ranges from 0 to 100.
On the weekly timeframe with a 14-period lookback (the standard), an RSI below 30 is considered oversold. For Bitcoin specifically, the weekly RSI falling below 30 has been an exceptionally rare event, occurring only during the deepest bear market capitulations.
Historical Evidence
| Date | Weekly RSI | BTC Price | Context |
|---|---|---|---|
| Jan 2015 | ~28 | ~$175 | Cycle 2 bottom |
| Dec 2018 | ~26 | ~$3,195 | Cycle 3 bottom |
| Mar 2020 | ~29 | ~$3,850 | COVID crash |
| Jun 2022 | ~27 | ~$18,000 | Post-Luna collapse |
| Nov 2022 | ~31 | ~$15,500 | Cycle 4 bottom (FTX) |
| Dec 2025 | ~35 | ~$80,500 | Current cycle correction |
| Feb 2026 | ~28 | ~$68,000 | Current |
As of February 2026, Bitcoin's weekly RSI has fallen to approximately 27.7 — the lowest reading since the $15,500 bottom in November 2022. Historical backtesting shows that 80% of instances where the weekly RSI fell below 36 led to higher prices within the same quarter.
Extreme RSI readings do not pinpoint exact bottoms. As analyst Lark Davis noted, "When BTC was this oversold in 2018, it dumped another 49%. In 2022, it dumped another 58%." RSI extremes are better understood as phase-shift signals — marking the transition from capitulation toward base-building — rather than precise turning points.
Sources: CryptoSlate (Feb 2026), TradingView/Cointelegraph (Dec 2025), MEXC analysis (Dec 2025)
Scoring
| Condition | Points |
|---|---|
| RSI < 30 (oversold) | +10 |
| RSI < 20 (extremely oversold) | +5 additional |
Implementation
The RSI period is adjustable (10–20 weeks), defaulting to 14.
Indicator 6: Drawdown from All-Time High
Theory
The simplest and most intuitive indicator: how far has Bitcoin fallen from its highest-ever price?
Every Bitcoin cycle bottom has been characterized by a severe drawdown. The pattern shows diminishing severity:
Historical Evidence
| Cycle | ATH | Bottom | Max Drawdown |
|---|---|---|---|
| 1 (2011) | $31.91 | $2.01 | -93.7% |
| 2 (2013–2015) | $1,177 | $175 | -85.1% |
| 3 (2017–2018) | $19,783 | $3,195 | -83.8% |
| 4 (2021–2022) | $69,045 | $15,500 | -77.6% |
| 5 (2025–??) | $126,296 | $68,000* | -46.2%* |
*Current cycle, ongoing.
The diminishing drawdown pattern reflects Bitcoin's maturation: deeper liquidity from institutional investors, spot ETFs, and broader market infrastructure cushion the downside in each successive cycle. If the diminishing trend continues, a 65–70% drawdown (implying a bottom around $39,000–$44,000) would fit the pattern, though the structural presence of ETFs and institutional buyers may compress this further.
Sources: Newhedge Bitcoin Drawdown, Galaxy Digital (Feb 2026), Binance analysis (Feb 2026)
Scoring
| Condition | Points |
|---|---|
| Drawdown > 70% | +10 |
| Drawdown > 80% | +5 additional |
Implementation
Indicator 7: Puell Multiple Proxy
Theory
The actual Puell Multiple, created by David Puell, compares the daily USD value of newly mined Bitcoin to its 365-day moving average. When miners earn far less than their yearly average (Puell < 0.5), they face financial stress and capitulation — historically coinciding with market bottoms.
Since this model doesn't have access to miner revenue data, it uses a proxy based on price returns:
When this proxy is deeply negative (below -30% to -50%), it indicates that recent returns are dramatically worse than the yearly average — a condition analogous to miner distress.
Historical Evidence
The true Puell Multiple has reliably identified cycle extremes:
- Puell > 4.0 (red zone): Has coincided with every cycle top
- Puell < 0.5 (green zone): Has coincided with every cycle bottom
- Puell ≈ 0.3: Historically marks the deepest bear market bottoms
At each halving, the Puell Multiple drops by approximately 50% (since block rewards are cut in half), mechanically pushing the metric toward the green zone. This creates a natural "reset" that often precedes the next bull run.
Sources: Bitcoin Magazine (Dec 2024), Newhedge Puell Multiple chart, Easy Crypto analysis
Scoring
| Condition | Points |
|---|---|
| Puell proxy < -30% | +5 |
| Puell proxy < -50% (severe distress) | +5 additional |
Composite Bottom Score
Design Philosophy
The composite score implements the principle of confluence — widely regarded in quantitative analysis as the most robust approach to signal generation. Rather than optimizing any single indicator, the model asks: how many independent analytical frameworks simultaneously suggest bottom conditions?
Source: Bitcoin Magazine Pro confluence analysis
Scoring Breakdown
| Indicator | Condition | Points |
|---|---|---|
| Price / 200 WMA | Ratio < 1.0 | +15 |
| Price / 200 WMA | Ratio < 0.7 | +15 |
| Pi Cycle Bottom | 21w MA below 67w MA | +15 |
| MVRV Z-Score Proxy | Z < 0 | +10 |
| MVRV Z-Score Proxy | Z < -0.5 | +10 |
| Log Regression Band | Price in lower band | +15 |
| Weekly RSI | RSI < 30 | +10 |
| Weekly RSI | RSI < 20 | +5 |
| Drawdown from ATH | > 70% | +10 |
| Drawdown from ATH | > 80% | +5 |
| Puell Proxy | < -30% | +5 |
| Puell Proxy | < -50% | +5 |
| Maximum possible | 120 (capped at 100) |
Weight Rationale
- 200 WMA (30 pts max): Highest weight because it has held as support in every single cycle and provides the most straightforward long-term valuation anchor.
- Pi Cycle (15 pts): Proven track record for tops (within 3 days); the bottom variant is less precise but captures the correct zone.
- Log Regression (15 pts): Mathematically elegant; lower band has captured every bear market low. Slightly uncertain due to ongoing refitting needs.
- MVRV Proxy (20 pts): The true MVRV Z-Score is arguably the single best cycle indicator; the proxy deserves meaningful weight despite being an approximation.
- RSI (15 pts): Strong historical correlation with bottoms, but can trigger early (months before the actual low).
- Drawdown (15 pts): Reliable but the diminishing-returns trend means the thresholds are shifting lower each cycle.
- Puell Proxy (10 pts): Lowest weight because the proxy is the loosest approximation; true Puell Multiple is far more precise.
Historical Validation
Backtested Composite Scores at Confirmed Bottoms
| Date | BTC Price | Composite Score | Indicators Active |
|---|---|---|---|
| Jan 2015 (~$175) | $175 | 85 | 200WMA ✓, Pi Cycle ✓, MVRV ✓, Log Band ✓, RSI ✓, Drawdown ✓ |
| Dec 2018 (~$3,195) | $3,195 | 80 | 200WMA ✓, Pi Cycle ✓, MVRV ✓, Log Band ✓, RSI ✓, Drawdown ✓ |
| Mar 2020 (~$3,850) | $3,850 | 70 | 200WMA ✓, Log Band ✓, RSI ✓, Drawdown partial |
| Nov 2022 (~$15,500) | $15,500 | 75 | 200WMA ✓, Pi Cycle ✓, MVRV ✓, RSI ✓, Drawdown ✓ |
False Positive Analysis
The model has produced scores of 40–60 during corrections that were NOT cycle bottoms (e.g., mid-2019 pullback, May 2021 crash). However, no non-bottom event has produced a score above 70, which is the threshold for the "Extreme Fear / Bottom Zone" classification.
Current Reading (Feb 27, 2026)
| Indicator | Value | Signal | Points |
|---|---|---|---|
| Price / 200 WMA | 1.153 | Neutral (above 200 WMA) | 0 |
| Pi Cycle Bottom | 0.917 (cross active) | Bottom signal active | 15 |
| MVRV Z-Score Proxy | 0.29 | Fair value | 0 |
| Log Regression Band | 6.57x lower band | Well above | 0 |
| Weekly RSI | 27.7 | Oversold | 10 |
| Drawdown from ATH | -45.4% | Correction, not capitulation | 0 |
| Puell Proxy | Mildly negative | No signal | 0 |
| Composite Score | 25/100 |
Assessment: The market is in a significant correction but has not reached the multi-indicator confluence historically associated with cycle bottoms. Only 2 of 7 indicators are active. For the composite to reach the 70+ "bottom zone," Bitcoin would likely need to decline further — potentially toward the $40,000–$55,000 range where the 200 WMA, realized price, and log regression lower band converge.
Limitations & Caveats
1. Past Performance ≠ Future Results
Bitcoin has only experienced 4 complete market cycles. A sample size of 4 is insufficient for statistical confidence. Any pattern derived from 4 data points could be coincidence.
2. Diminishing Returns Are Changing the Rules
As documented by analysts like Benjamin Cowen, diminishing returns and compressed volatility are causing many traditional indicators to trigger earlier or at different thresholds than they did historically. The Pi Cycle bottom indicator, for example, fired 5 months early in 2021. The 200-week SMA failed to mark the bottom cleanly in 2023.
3. Structural Market Changes
The introduction of spot Bitcoin ETFs in January 2024, which have accumulated over $54 billion in net inflows, has fundamentally altered Bitcoin's market structure. ETF flows amplify both upside and downside moves, and institutional holders may behave differently from retail-driven cycles. The current cycle may not follow historical patterns.
4. Proxy Limitations
The MVRV Z-Score proxy and Puell Multiple proxy are mathematical approximations of on-chain metrics. They correlate directionally but may diverge significantly from the true values during periods of unusual on-chain activity (e.g., large exchange movements, miner capitulation events, lost coins).
5. No Timing Precision
The model identifies zones of bottom probability, not exact dates or prices. Historical bottoms have formed over periods of 2–8 months within the high-probability zone. Even a score of 90+ does not mean the absolute bottom has been reached — further downside is always possible.
6. External Risks Not Captured
The model is purely technical and does not account for macro risks (interest rates, tariffs, regulatory changes), black swan events (exchange collapses, protocol bugs), or structural shifts in Bitcoin's monetary policy. The FTX collapse in November 2022, for example, extended the bottom beyond what any technical model predicted.
This model is for educational and analytical purposes only. It does not constitute financial advice. Bitcoin is a highly volatile asset, and past patterns may not repeat. Always conduct your own research and consult a financial advisor before making investment decisions.
Sources & References
- 200-Week Moving Average: Bitcoin Magazine / Nasdaq — "Forecasting Bitcoin Price Cycle Peak with the 200-Week Moving Average" (Jun 2025)
- Pi Cycle Top Indicator: Bitcoin Magazine Pro — Pi Cycle Top Indicator (Philip Swift, Apr 2019)
- Pi Cycle Accuracy: Nasdaq — "The Bitcoin Pi Cycle Top Indicator: How to Accurately Time Market Cycle Peaks" (Nov 2024)
- MVRV Z-Score: Bitcoin Magazine Pro — MVRV Z-Score Chart (Original creators: Murad Mahmudov & David Puell)
- MVRV Z-Score (Glassnode): Glassnode Studio — MVRV Z-Score
- MVRV Analysis: Bitcoin Magazine — "Mapping Bitcoin's Bull Cycle Potential" (Jun 2025)
- Current MVRV Reading: AInvest — "Market Sentiment Indicates Distance From Typical Bottoming-Out Region" (Feb 2026)
- Logarithmic Regression Bands: TradingView — CryptoRiskAlert indicator (2019/2021)
- Logarithmic Regression Analysis: Benjamin Cowen — "Bitcoin Peak Logarithmic Regression Band" (Mar 2021)
- Weekly RSI Analysis: CryptoSlate — "Bitcoin is currently oversold more than any time in history" (Feb 2026)
- RSI Historical Correlation: MEXC — "Bitcoin Weekly RSI Nears Historic Lows at $90K" (Dec 2025)
- Drawdown Analysis: Galaxy Digital — "Bitcoin Drawdown Nears 40%" (Feb 2026)
- Drawdown Historical Data: Newhedge — Bitcoin Price Drawdown from ATH Chart
- Puell Multiple: Bitcoin Magazine — "What Is The Bitcoin Puell Multiple Indicator" (Dec 2024)
- Confluence Methodology: Bitcoin Magazine Pro — "Mastering Confluence For Superior Bitcoin Signals"
- 4-Year Cycle Evidence: Kaiko Research via BeInCrypto/Yahoo Finance — "Bitcoin's Four-Year Cycle Is Intact" (Feb 2026)
- Halving Cycle Timing: Bitcoin Suisse — "The Bitcoin Halving and the Bitcoin Market Cycle"
- 200 WMA Break: TradingView/Cointelegraph — "Critical Bitcoin Weekly Trend Breaks" (Feb 2026)
- Multi-Indicator Methodology: Binance Square — "10 Proven Indicators to Predict Bitcoin's Next Top and Bottom" (Sep 2024)
- On-Chain Bottom Identification: CoinDesk/Yahoo Finance — "This onchain metric has identified the Bitcoin bottom every cycle" (Feb 2026)