Ghost Alpha

← Learn

Halving Cycle Position

Where today sits between the last Bitcoin block-reward halving and the estimated next one - the classifier's own point-in-time position within the roughly four-year issuance schedule, used alongside valuation, flows, and leverage, never on its own.

Halving Cycle Position

What it measures

Bitcoin's block reward halves roughly every four years (every 210,000 blocks). This measures where "today" sits within that schedule: the number of days since the last halving, and an estimated percentage of the way through to the next one. The last halving's date is a confirmed historical fact; the next halving's date is an estimate that assumes Bitcoin's roughly 10-minute average block time holds, and every field derived from it is labeled as an estimate rather than presented as confirmed.

Formula

Cycle Progress % = Days Since Last Halving / (Estimated Next Halving Date - Last Halving Date) × 100

The classifier further buckets this progress percentage into three named phases - early, mid, late - using the same thresholds recorded in /methodology/read/.

Normal range

By construction, progress runs from 0% (the day of a halving) to 100% (the day of the next one), then resets. There's no "normal" level within that range - it's a calendar position, not a market reading.

How it fails

The calendar doesn't know anything about market conditions. Two points at the same cycle progress percentage, years apart, can sit in completely different valuation, flow, and leverage environments - which is exactly why halving position is one input alongside those others, never the anchor.

One input, not the anchor

A growing share of serious commentary argues the four-year halving cycle itself has been dampened, or broken outright, by spot ETF flows and growing institutional ownership changing who holds bitcoin and why. This project's identity can't be hostage to that argument turning out to be right. That's why the regime read is built on observed state - valuation, flows, leverage, and sentiment - with halving position folded in as one input among several, not the organizing idea. If the calendar cycle stops mattering, the classifier keeps working, because it was never built to depend on it alone.

Related metrics

MVRV, Puell Multiple

Bull read

Being early in a cycle has historically coincided with lower valuation readings and more room before the kind of stretched conditions that have preceded past tops - but this reading only means something combined with what valuation and flows are actually showing today, not on its own.

Bear read

Being late in a cycle doesn't mechanically force a top on any fixed date - and if the four-year pattern really has been dampened by ETF-era ownership, "late in the cycle" may end up meaning less than it has in prior cycles.

Worked example

No worked example. This metric has no stored daily history to compute a worked example from.

Full metric page →