Photorealistic Bitcoin coin surrounded by an orbital ring of glowing segments -- most in neutral white, one in orange as the current active point, and the last one a fading, incomplete dashed arc, over a dark background with reflected candlesticks
Original diagram illustrating this page's version timeline -- not a capture of any tool on the site.

Version History of the BlockPulse Analytics Score

The Score's formula hasn't changed since its launch on July 1, 2026. This page exists so that can be checked with dates, not our word: any future change, to composition or formula, will show up here first.

Every market indicator carries the same reasonable suspicion: "surely they tweak it after seeing what result suits them." The only honest way to answer that isn't to promise it never happens -- it's to leave a dated trail of every real change, and of every change that was evaluated and not adopted, which is usually the part nobody shows.

If you don't yet know how the Score is calculated, the full methodology is the best place to start; this page assumes you already know it and focuses only on its change history.

Prehistory: the previous method (June 2026)

Before today's Score there was a majority-vote zone indicator: nine market signals, each weighted equally, deciding between accumulation and distribution by simple majority. It wasn't removed when the weighted Score launched -- it's still visible on the own-node page as "Previous method: majority vote," precisely so the before and after can be compared with real data, not from memory.

v1 -- the current weighted Score (July 1, 2026)

On July 1, 2026, the weighted verdict that's still in production today was born: instead of an equal vote, each indicator contributes a score from -100 to +100 with its own weight, and the result is their weighted average. That same day, global liquidity was added as a tenth indicator -- the only composition change the Score has had since. This is the version we call v1, and it's the one still running live right now.

Reserve policy (July 5, 2026)

Four days after launching v1, an explicit decision was made: stop publishing each indicator's exact weight and contribution magnitudes, both on the site and in the API. What's still published in detail is what makes up the Score, what each piece measures, and which direction it's currently pushing the result; what's withheld is the how much.

The justification isn't hiding for its own sake: the legitimacy of that reserve comes from being able to check the Score's public track record (see below), not from having to trust our word about the formula. And there's an honesty nuance that's part of this version too: the weights in effect until that date were exposed unfiltered in the API, so they're treated as potentially known by anyone who could have checked them before July 5. The reserve protects future revisions of the Score, it doesn't erase what could already be seen before that date.

Public backtest (July 6, 2026)

A day later, the Market section published the Score's historical reconstruction from 2018 to today, overlaid on Bitcoin's price -- so anyone can check what the Score would have shown at any point in the cycle, without having to trust a summary. That reconstruction uses 7 of the live Score's 10 indicators (without our own node, which has no continuous history going back), and that's why its average confidence sits around 60% versus the live Score's usual ~95% -- a lower figure that's the honest signal that fewer pieces are being used, not a calculation error. You can read more about its real limits in where the Score has failed.

Change evaluation protocol (July 8-11, 2026)

From this date on, any change proposed for the Score -- a new indicator, a threshold adjustment, a different way of combining signals -- goes through an internal protocol before it can be adopted: it's calibrated on one stretch of history and validated on a completely different one, never the same one, and it's only adopted if it improves things robustly in both stretches at once, not just in the aggregate. A dated record is kept of every variant evaluated, what it showed, and why it was adopted or discarded.

This is the central fact of this page: since the protocol has existed, close to 10 distinct improvement variants have been evaluated, spread across 3 rounds of experiments separated by date, and none has met the bar for adoption. The Score's live formula remains exactly the same as it was on July 1, 2026. What was tested and how each variant works isn't named here -- that detail is part of what's withheld -- but the existence of the three rounds, their dates, and their result (zero adoptions) are public on purpose: it's the proof that evaluating isn't the same as adjusting.

Live history and vintage snapshots (since July 8, 2026)

Since July 8, 2026 -- a week after v1 launched -- the values of the indicators that feed the Score have been stored daily with their exact capture date -- a "vintage" snapshot of the data as it was known that day. That lets us recalculate the Score for any future day from frozen data, and makes that history impossible to edit with the passage of time: reconstructing the past with today's data (as the 2018-2026 backtest does) isn't the same as freezing each day as it was seen at the time. It's the difference between asking for trust and allowing it to be checked.

v2 -- announced, no date

Once our own Bitcoin node finishes syncing, the Score's heaviest-weighted indicator will be able to run on complete data instead of partial, and that will be the occasion to re-evaluate the whole Score under the same protocol above -- not before. There's no date yet because it depends on how long syncing takes, not on a product decision.

This page's commitment is simple and forward-looking: no change to the Score's formula will be adopted without an entry here, with its date and its reasoning -- whenever v2 ships, or if any variant currently under evaluation ever meets the criterion before that.

You can see the Score live, with all this context, in the Market section.

Last updated: 2026-07-10