Rotation Signal
SOURCE ERROR

Data & Methodology

This model is rules-based and fully auditable. Every reading on the dashboard can be traced back to the inputs, weights and thresholds documented here. Nothing on this site predicts price or recommends a trade.

What the four numbers mean

Rotation stage (0–6)
A discrete description of where liquidity currently sits. Stages only change when conditions persist, so the reading does not flip on a single volatile day.
Rotation strength (0–100)
How much independent evidence supports capital rotating outward from Bitcoin right now.
Signal confidence (0–100)
How trustworthy that reading is: data freshness, persistence of conditions, and whether independent signals agree. Strong rotation on low confidence deserves scepticism.
Market heat (0–100)
How stretched and speculative the market is. High heat increases risk. It is never treated as bullish confirmation.

The model is universal, not personal

The market reading is calculated from the whole market. Its building blocks are Bitcoin, Ethereum, and three market segments — large cap, mid cap and higher beta — whose membership is set purely by market-capitalisation rank across the top 250 assets. No individual coin is written into the model, and segment membership changes automatically as the market changes.

Watchlists are an overlay. Whatever you add to a watchlist changes only what you are shown about those assets; it never affects the rotation stage, strength, confidence or heat. Two people with completely different watchlists see the same market reading.

Where an input is missing or a source fails, the affected measure is reported as unavailable and confidence is reduced. Missing data is never filled in with an estimate.

Data sources

  • CoinGecko

    Prices, market capitalisation, volume, daily history, Bitcoin dominance and total market capitalisation.

    UNKNOWN

  • Alternative.me

    Crypto Fear & Greed Index, daily.

    UNKNOWN

  • DefiLlama

    Aggregate stablecoin capitalisation history.

    UNKNOWN

All external requests are made on the server, batched, and cached. The model refreshes on a 30-minute cycle, which keeps usage well inside free public limits. Freshness is labelled LIVE (under 15 minutes), CURRENT (under 45), DELAYED (under 2 hours), STALE (older), or SOURCE ERROR when a provider fails.

Scoring weights

  • Bitcoin health12%

    Is Bitcoin itself in a healthy trend? Rotation from a broken Bitcoin is distribution, not rotation.

  • Bitcoin dominance18%

    Bitcoin's share of market value, gated against total market direction so a falling share only counts when capital is not fleeing.

  • ETH/BTC relative strength22%

    The single most important early rotation tell. Requires persistence, not a one-day move.

  • Large-cap segment16%

    The largest non-Bitcoin, non-Ethereum assets by market value, measured as a group. Membership is set by market-cap rank, never by name.

  • Mid-cap segment12%

    The next tier of the market. Broadening here is what separates a real rotation from a two-asset move.

  • Altcoin breadth12%

    How widely strength is shared across the top 50, excluding stablecoins and wrapped duplicates.

  • Liquidity8%

    Total market capitalisation and stablecoin supply — is new capital actually arriving?

Stage definitions

  1. 0

    Risk off

    Bitcoin below trend and capital contracting. Rotation questions do not apply.

  2. 1

    Bitcoin accumulation

    Bitcoin stabilising and reclaiming trend while alternatives lag.

  3. 2

    Bitcoin leadership

    Bitcoin strong and dominance rising. A healthy market, but not a rotation market.

  4. 3

    ETH rotation

    ETH/BTC strength persists and dominance falls while the total market holds up.

  5. 4

    Alt rotation confirmed

    The large-cap and mid-cap segments join, and breadth confirms participation is real.

  6. 5

    High-beta expansion

    The higher-beta segment leads. Returns can be large and drawdowns severe.

  7. 6

    Speculative euphoria

    Heat, sentiment and breadth are all stretched. Treated as elevated risk, never as confirmation.

Known limitations

  • · Historical rotation patterns can break. The model describes present conditions; it does not forecast.
  • · Free public data sources can be delayed, revised, or briefly unavailable. Status labels always show when this happens.
  • · Market-structure history collected before this model began running is reconstructed from a large-cap basket and is an approximation. Readings recorded live are measured directly.
  • · Breadth is measured across the top 50 assets excluding stablecoins, wrapped tokens and staking derivatives, which is a deliberate simplification of the wider market.
  • · Every input is stored alongside each reading, so past readings can be re-examined and the model can be revised transparently under a new version number.