Cycle Phase ProbabilityTRINITY EXCLUSIVE

Trinity exclusive model

This metric is a proprietary Trinity Insights model. Its formula, inputs, weights and parameters are NOT disclosed. The page documents only the output (bounded scale, interpretation zones, historical context). Access to the score and its time series is via the REST API and the MCP server, subject to the required tier.

What is it?

Most cycle indicators give you a single answer: 'we are in the bull phase' or 'the bear is starting.' But the real world is not binary. The Cycle Phase Probability chart provides something far more honest: a real-time probability distribution across four possible phases - Accumulation, Markup (Bull), Distribution, and Markdown (Bear). Instead of telling you a definitive answer, it tells you the odds. When the Accumulation probability is at 75% and Markup is at 20%, the model is saying 'we are almost certainly in accumulation, with a small chance the bull has already started.' This probabilistic framework prevents the overconfidence that deterministic models encourage and provides a more nuanced view of cycle positioning that sophisticated users and fund managers require.

How to read

The chart displays four stacked areas that always sum to 100%. Each color represents one phase: green (Accumulation), blue (Markup/Bull), red (Distribution), and gray (Markdown/Bear). When one phase dominates (occupying >70% of the area), the model has high confidence in that phase classification. When the probabilities are roughly equal (each around 25%), the model is uncertain - typically during phase transitions. The BTC price is overlaid on a secondary axis. The transitions between dominant phases are the key moments: watch for the blue (Markup) area expanding at the expense of green (Accumulation) - this transition has historically coincided with the beginning of the most powerful rally phase.

Key zones

Phase probability patterns at historical inflection points: • Cycle bottoms: Markdown probability typically peaks at 85-95% during the final capitulation, then rapidly declines as Accumulation rises. The crossover (Accumulation > Markdown) has occurred within 30-60 days of the confirmed bottom. • Bull confirmation: When Markup probability exceeds 50% for the first time, the bull phase has historically been confirmed. This has occurred approximately 4-6 months after the cycle bottom. • Cycle top warning: Distribution probability begins rising from <5% to >20% approximately 2-3 months before the cycle top. When it reaches 40%, the top has typically already occurred or is imminent. • Phase transition uncertainty: During the 2-4 week periods where no single phase exceeds 40%, the model is genuinely uncertain. These windows have historically preceded significant volatility. • The 'clean' transitions: The sharpest, most profitable moves have occurred when the model transitions rapidly from one dominant phase to another (e.g., Accumulation at 80% dropping to 20% as Markup rises to 70% within 2 weeks).

What to observe

• Probability concentration: When any single phase exceeds 80%, the model has high conviction. These high-conviction periods have historically been the clearest cycle positioning indications. • Early Distribution rise: The Distribution probability rising above 15% while Markup is still dominant has been an early warning of cycle top approach - typically 2-4 months before the peak. • Accumulation-Markup transition: This is the single most profitable cycle event. When the blue area begins growing rapidly at the expense of green, the bull phase is likely starting. • Markdown duration: The Markdown phase probability typically dominates for 10-14 months during bear markets. When it begins declining below 50%, the bear may be approaching its end. • Model uncertainty as information: Uncertainty (no dominant phase) is itself an indication - it indicates a regime transition is underway. The direction of resolution provides the next major indication.

Historical context

The Cycle Phase Probability model was developed to address a fundamental limitation of traditional cycle analysis: the tendency to force binary classifications on inherently uncertain situations. Traditional approaches say 'we are in the bull phase' or 'the bear has started,' but the reality is that phase transitions unfold over weeks or months, with periods of genuine ambiguity. The probabilistic approach was inspired by regime-switching models in econometrics (Hamilton, 1989) and Hidden Markov Models in machine learning. The model uses multiple independent indication inputs to estimate the likelihood of each phase, producing a distribution that honestly reflects the current state of evidence. This approach has proven especially valuable during the 2021 double-top period, where the model correctly showed Distribution probability rising between the April and November peaks - providing warning even as price remained elevated.

Expert notes

⚠️ Trinity Exclusive Model - The model uses a modified Hidden Markov Model (HMM) with 4 hidden states (Accumulation, Markup, Distribution, Markdown) and 8 observable features (the same sub-indicators that feed the TCCI). The emission probabilities are estimated using Gaussian mixture models fitted to historical data for each phase. Transition probabilities are informed by the historical phase sequence (A→M→D→B→A) and the time elapsed in each phase relative to historical averages. The output is the posterior probability of each state given all observations up to the current day, computed via the forward algorithm. The model is re-estimated weekly using a rolling 8-year window to adapt to structural changes. Backtesting with walk-forward validation shows the model correctly identifies the dominant phase (>50% probability for the actual phase) approximately 82% of the time across the 2012-2025 period. The specific combination of HMM architecture, feature set, and adaptive re-estimation is proprietary - no equivalent industry indicator exists.

Common mistakes to avoid

• Treating probabilities as certainties: A 70% Markup probability means there is still a 30% chance the model is wrong. Size positions accordingly. • Assuming phases must follow the canonical sequence: While A→M→D→B is the typical order, the model can show phase skipping (direct A→D) in unusual market conditions. • Reacting to the minority phase: If Accumulation is at 75% and Distribution is at 5%, do not anchor on the 5%. Focus on the dominant phase and monitor for change. • Expecting precise transition timing: Phase transitions can take 2-8 weeks to complete. The model's transition is gradual, not instantaneous. • Comparing probabilities across cycles: The same probability level (e.g., 70% Markup) can correspond to different price environments in different cycles. Use the probability for phase identification, not price prediction.

Programmatic access

REST API

curl -sS \
  'https://api.trinityinsights.io/api/v1/cycle-intelligence/cycle-phase-probability/history?days=90' \
  -H 'X-API-Key: $TRINITY_API_KEY'

MCP server

{
  "tool": "get_chart_value",
  "metric_id": "cycle-phase-probability",
  "timeframe": "1y"
}

Required tier: performance. See the pricing grid for the tier list and the MCP documentation for multi-client configuration.

Institutional disclaimer

Trinity Insights is an educational and analytical tool. The metric above does not constitute investment advice. Trinity Insights is not a Crypto-Asset Service Provider (CASP) registered under MiCA Regulation (EU) 2023/1114. See the full disclaimer.