Bitcoin Cycle Tops and Bottoms: The Pattern — and Where Our Score Missed It

Bitcoin cycle tops and bottoms haven't lined up with the same signals by accident. Here they are with the exact thresholds, the verified November 2022 case, and the cycle extremes our own combined signal has missed, documented, not hidden.

MVRV Z-Score, the Pi Cycle indicator, exchange reserve trackers: Bitcoin's on-chain corner has no shortage of tools claiming to spot cycle turns. What's rarer is a source that also publishes the times its own combined read got it wrong. This piece does both: it documents the pattern that has repeated at every past top and bottom, with the exact thresholds and the FTX-collapse case already verified with data, and it links straight to the cycles where our own Score missed the call entirely, because a track record only means something if the misses are on the record too.

What has repeated at Bitcoin cycle tops

At past Bitcoin cycle tops, the following have repeatedly coincided:

• MVRV above 3: the market as a whole sitting on a large unrealized gain, though that threshold gets harder to reach every cycle, see below.
• Puell Multiple above 4: miners earning far more than usual, a sign price has risen sharply in a short window.
• Extreme euphoria in sentiment: the fear and greed index at its highest zone, sustained for days, not a single spike.
• Price well above its 200-day moving average: a fast rise relative to its own recent trend, not just a high price in absolute terms.

None of these alone is enough: what has marked a top in past cycles is several coinciding at once, not one spiking while the rest sit at normal levels.

What has repeated at cycle bottoms, and the verified November 2022 case

At the other end, past cycle bottoms have coincided with the exact opposite: MVRV at or below 1 (the market underwater on paper), Puell Multiple below 0.5 (miner capitulation, see capitulation signals for how it's measured), sustained extreme panic in sentiment, and price well below its 200-day average.

The clearest verified case is November 2022. After FTX collapsed, several of these bottom signals coincided at once. In our historical reconstruction, the Score reached +51.1 ("Moderate Accumulation"), its highest reading across 8.5 years of reconstruction, while the dominant narrative was that Bitcoin would never recover. With the information available today, it was a real cycle bottom. Worth the honest caveat up front too: even at its best moment, the Score has never once reached the "Strong Accumulation" band (+65) in the full reconstruction, detailed, unedited, in where the Score has failed.

Why a fixed threshold expires every cycle

Documenting that these signals coincided in the past isn't the same as guaranteeing they'll line up exactly the same way next time, and there's a concrete reason, not just generic caution. In a review of the last five Bitcoin cycle tops, the peak MVRV reached at each one was, in order:

7.74 → 5.25 → 4.72 → 2.85 → 2.29

Each successive top landed lower than the one before it (more detail in the MVRV article). The "above 3" used above as the euphoria reference would have been conservative in the first cycle of that series, and is already hard to reach in the most recent one: a 2.85 MVRV never crossed it, despite marking a real top all the same. Puell Multiple follows a similar drift, though it isn't tracked with the same precision yet. It's the same reason a normalized variant, the MVRV Z-Score, exists in the first place: to stop comparing cycles against a fixed number that keeps expiring.

Only 3-4 complete cycles: the honest limit of the sample

Bitcoin has only had 3-4 complete cycles to date, too small a sample by any serious statistical standard, for the thresholds above or for the Score itself. Each historical top and bottom also had its own quirks that won't repeat exactly. The honest position isn't promising the next cycle will follow the same script precisely; it's documenting what can be checked today and correcting the model when new data contradicts it, exactly what where the Score has failed does, in detail, without softening the numbers.

Where our own Score has missed these turns

None of the trackers running MVRV Z-Score dashboards or Pi Cycle countdowns disclose when their own read got a major extreme wrong. Ours does, on purpose: the reconstructed Score stayed flat in the "Neutral" zone through the December 2018 bottom, the March 2020 COVID crash, and the October 2025 top: three of Bitcoin's most extreme moments, all missed. That's not a footnote we buried; it has its own page, with the exact numbers for each miss.

The pattern itself isn't wrong in those three cases. What's missing is that no single indicator crossed its own extreme threshold hard enough, at the same time as the others, for the combined read to register anything but Neutral. A pattern documented from a handful of past cycles will keep having blind spots like that; publishing them is the only honest way to let anyone judge how much weight the pattern deserves going forward, rather than taking the two verified hits on faith and assuming the misses don't exist.

See the Score's full history overlaid on price, with cycle tops and bottoms marked, on the live Score page, free, no sign-up, public methodology.

Last updated: August 31, 2026