Cycle Convergence Framework
CCF historical backtest
Forward outcome context when five pillars converged at confirmed cycle extremes since 2015.
Quick answer
The Cycle Convergence Framework backtest built a 0-100 composite from RSI, EMA gap, Bollinger %, drawdown, vs-200MA, and MVRV. When score exceeded 60 at confirmed extremes: bottoms were positive at 90 days 58% of the time (42% false positive rate). Tops saw lower prices at 180 days 81% of the time (19% false positive rate). Past patterns, not forecasts.
Bottom hit rate
58%
Positive at 90 days
Top hit rate
81%
Lower at 180 days
Top false positive
19%
vs 42% bottoms
- CCF composite score
- A 0-100 index built from six inputs: RSI, EMA 10/100 gap, Bollinger Band %, drawdown from 1-year high, distance from 200-day MA, and MVRV. Score above 60 at confirmed cycle extremes was tested against forward returns at 30, 90, 180, and 365 days.
Bottom Zone
3+ pillars vote cycle-low
Mid-Cycle
Mixed / neutral votes
Top Zone
3+ pillars vote cycle-high
How the CCF framework was backtested
The FynSight research team started with scipy local extrema detection on Bitcoin daily data since 2015. Nine confirmed bottoms and nine confirmed tops were labeled. For each extreme, indicator readings at the turn were recorded. Thresholds were calibrated so pillar votes matched historical behavior at those points.
A separate backtest script built a 0-100 composite score from six continuous inputs. When the score crossed 60 near a labeled extreme, forward returns were measured at 30, 90, 180, and 365 days. The goal was not to optimize for maximum profit. It was to measure how often convergence resembled prior cycle zones where forward outcomes skewed positive (bottoms) or negative (tops).
FynSight provides market observations only. Not a SEBI-registered investment advisor. No buy, sell, or hold language. Zones are context tools, not trade signals.
Full Cycle Convergence Framework methodology →What goes into the composite score
Six inputs, each normalized to contribute to a 0-100 scale: RSI (price exhaustion), EMA 10/100 gap (trend structure), Bollinger Band percentage (statistical stretch), drawdown from 1-year high (cycle depth), distance from 200-day moving average (trend anchor), and MVRV (on-chain valuation).
The live five-pillar vote system uses binary thresholds derived from this research. The composite score is the continuous version used for backtesting granularity. Score above 60 roughly corresponds to three or more pillars voting in the same direction with moderate strength.
Power Signal upgrade (MVRV + RSI together) maps to the highest composite readings in the historical sample. Those readings had the cleanest forward outcome separation.
Bottom zone backtest results
Bottom score > 60
- 58% positive at 90 days
- Avg forward return: +47% in 90 days
- False positive rate: 42%
Fear is chaotic. Bottoms grind or crash fast. Tops cluster cleanly when euphoria is uniform.
Top score > 60
- 81% lower at 180 days
- Avg drawdown: -34% in 180 days
- False positive rate: 19%
Top signals are more reliable than bottom signals in the historical sample.
Bottom threshold: composite score above 60 near a cycle-low label. 58% of cases showed positive returns at 90 days. Average forward return: +47%. False positive rate: 42%.
Fear is chaotic. Bottoms grind or crash fast. Tops cluster cleanly when euphoria is uniform. Dec 2018 and Nov 2022 both scored high and delivered 120%+ 1-year returns. Mid-cycle dips like Jul 2021 scored high on RSI and depth but failed on MVRV, and correctly would not have met full convergence in the five-pillar system.
| Date | Price | RSI | MVRV | Notes |
|---|---|---|---|---|
| Aug 2015 | $210 | 23.3 | — | — |
| Jun 2018 | $5,903 | 26.6 | — | -18.9% |
| Dec 2018 | $3,237 | 26.1 | 0.69 | -36.0% · -83.9% · +121% |
| Mar 2020 | $4,971 | 6.6 | — | -64.0% |
| Jul 2021 | $29,807 | 23.7 | 3.36 | -18.9% · -54.0% |
| Nov 2022 | $15,787 | 25 | 0.75 | -16.4% · -73.4% · +127% |
| Sep 2024 | $53,949 | 26 | — | -6.1% |
| Apr 2025 | $76,272 | 26.6 | — | -8.2% · -12.2% |
| Jun 2026 | $60,867 | 6.2 | 1.135 | -51.8% |
Top zone backtest results
Top threshold: composite score above 60 near a cycle-high label. 81% of cases saw lower prices within 180 days. Average drawdown: -34%. False positive rate: 19%.
Top signals are more reliable than bottom signals in the historical sample. Oct 2025 at $124,753 is the most recent high-score top cluster. Five months later, price was down approximately 47%.
| Date | Price | RSI | Notes |
|---|---|---|---|
| Dec 2015 | $465 | 81.1 | — |
| Jun 2016 | $766 | 89.4 | +35.1% |
| Dec 2017 | $19,497 | 78.7 | +93.6% |
| Jun 2019 | $13,016 | 95 | — |
| Feb 2020 | $10,326 | 75.8 | — |
| Apr 2021 | $63,503 | 75.3 | — |
| Nov 2021 | $67,567 | 75.2 | — |
| Mar 2024 | $73,084 | 80.1 | +37.7% |
| Oct 2025 | $124,753 | 83.4 | — |
Why top signals outperform bottom signals statistically
Greed clusters. Fear disperses. At cycle tops, retail, institutions, and leveraged traders tend to agree the trend is unstoppable within the same few weeks. RSI, MVRV, funding, and EMA gap spike together. The composite score fires cleanly.
At bottoms, capitulation happens in waves. RSI can exhaust in June, again in September, and finally at the true low in November. MVRV may not cross 1.0 until the last wave. Funding flips negative late. The composite score triggers more false positives because inputs do not align on the same schedule.
Practical implication: top zones are more actionable for risk reduction (trimming leverage, partial profit-taking). Bottom zones are more actionable for accumulation context but with lower statistical confidence. This matches how searchers use the data: tops for protection, bottoms for patience.
Practical top zone guide →Case studies from the calibration set
Dec 2018 generational bottom
BTC $3,237. RSI 26.1. MVRV 0.69. EMA gap -36.0%. Drawdown -83.9%. Full convergence. Forward 1-year: +121%. Textbook high-conviction bottom zone in hindsight.
Jul 2021 false bottom
BTC $29,807. RSI 23.7. MVRV 3.36. Drawdown -54%. Partial convergence only. Forward path led lower to Nov 2022. MVRV filter prevented treating this as generational.
Oct 2025 distribution top
BTC $124,753. RSI 83.4. MVRV 2.29. Multiple pillars elevated. Forward 5 months: -47%. Convergence visible before mainstream top calls.
Compare to live readings today →Backtest limitations every reader should know
Sample size is small: nine bottoms and nine tops since 2015. Bitcoin's market structure changed with spot ETFs, institutional custody, and different leverage venues. Past zone behavior is historical context, not a guarantee.
The backtest measures proximity to labeled extremes, not optimal entry or exit timing within a zone. Zones can last weeks or months. Forward return windows (90d, 180d) are arbitrary cutoffs chosen for comparability.
CCF publishes transparent scorecards on the crypto calibration page so readers can see how similar live setups resolved in the historical sample without treating backtest stats as predictions.
What people search about Bitcoin cycle backtests
Common queries: “bitcoin cycle indicator accuracy”, “onchain indicators backtest”, “bitcoin bottom signal success rate”, “has bitcoin topped indicators”. This page is the canonical answer within FynSight for CCF accuracy stats.
External frameworks (CBBI, Bitcoin Barometer, BCI) publish similar convergence models. CCF differentiates by publishing pillar-level thresholds, full historical tables, and live votes on one interconnected cluster rather than a single composite number without context.
Live calibration and scorecards
Theory without live data is incomplete. The cycle position page shows current pillar votes. The crypto calibration page shows how often similar composite readings resolved in the historical sample.
Frequently asked questions
How was the CCF framework backtested?
A 0-100 composite score was built from RSI, EMA gap, Bollinger %, drawdown, distance from 200-day MA, and MVRV. Threshold crossings at confirmed cycle extremes were tested against forward returns at 30, 90, 180, and 365 days.
Are Bitcoin top signals more accurate than bottom signals in CCF?
Yes in this dataset. Top score above 60: 81% saw lower prices at 180 days, 19% false positive rate. Bottom score above 60: 58% positive at 90 days, 42% false positive rate.
What is a false positive in the CCF backtest?
A high composite score that did not lead to the expected forward outcome. Bottom false positives are common because fear disperses gradually. Top false positives are rarer because greed clusters.
Does the backtest predict future Bitcoin prices?
No. It measures how often historical zone conditions resembled prior extremes where forward outcomes skewed in one direction. Past patterns, not forecasts.
Related guides in this cluster
Common searches about Bitcoin cycle tops and bottoms, mapped to FynSight CCF pages.
Explore the CCF cluster
FynSight provides market observations only. Not a SEBI-registered investment advisor. No buy, sell, or hold language. Zones are context tools, not trade signals.