Training note, written on 25 September 2026 from data as of 5 July 2026. Not part of the published series.

Fewer Commits And A Rising Price At Ether.fi

By the close of 5 July, the ETHFI token was climbing, and the code work on Ether.fi was running well behind its peers. For anyone holding ETHFI, the price and the only measure of the project's own work point opposite ways. Both sit on the cover.

Only one bar appears there: developer commits, the code changes logged each week in the project's public repositories, 30.2 points below the median of its group. One bar, one measure. Ether.fi is a liquid staking protocol on Ethereum, a service that stakes ether for its users, and ETHFI is its token.

Counting the last four complete weeks against the four before, commits fell 56.8%, from 347 to 150. The 10 members of the 30-strong Liquid Staking group that report commits, Ether.fi included, had a median change of -26.6%. The commits turned first, in the week of 31 May to 6 June, and fell in 2 of the 4 weeks that followed, the four whose total is down 56.8%. The price turned up on 3 July.

Here is how I will know whether the gap closes. If developer commits (as totals of 4 complete weeks) grow at least as fast as the median of their group in the 30 days to 4 August, the gap this note is about has closed. This note follows development shrinking while the token rises; commits keeping pace with their group would remove the shrinking half. The price alone would be left.

ETHFI stood at $0.4112 on 5 July, up 9.8% that day alone. Over 30 days, end against start, it gained 28.1%, while the median of the 6 group members with a listed price rose 13.3%. That is 14.8 points ahead. Nothing here ties commits to the price.

Any skeptic starts with the thinness: one measure, whose weekly count has ranged between 3 and 102 over the last 12 complete weeks (the latest, 29, sits in between). That point stands. Of usage, capital and development, read apart, only development is in these numbers for Ether.fi: no fees, trading volume, value locked or user counts, no news or reasons, nothing after 5 July.

Against that, net of peers, this 30-day change had a middle half of +9.8 to +149.2 points over 90 days; -30.2 is below. The group is 30.2 points ahead. That is the only reason I give one measure any weight, and one measure is a thin reason. On the missing measures the skeptic wins: I have no clean way to judge a protocol on code alone.

Nothing here says what that code was meant to build, the thing I would most want to know before judging who is right. Commits are counted here, not read.

Sources: GitHub (developer commits), CoinGecko (price).

Peer group: 32 liquid staking protocols tracked by DefiLlama (who they are). 'Net of the group' is the 30-day change minus the group's median.

Not financial advice. This is a research note based on public on-chain data.