Njal
@cryptonjal
Got some interesting replies, and tried to test some hypothesis with data (other replies were interesting as well, but not tested by me). 🧵 First, @mvr pointed out to zoom out. - The downtrend of Jan. and Mar. '25 is part of a larger downtrend that's started on May 22nd, 2024. - The uptrend from Feb till end May was (imo) driven by the innovation of frames, and degen tipping. Interestingly enough also on May 22nd, 2024 there was the news that Farcaster raised $150M. https://cointelegraph.com/news/degen-token-surges-17-farcaster-raise-150-million The Theory of Contrary Opinion would suggest that such good news is always around the top.
3 replies
2 recasts
3 reactions
Njal
@cryptonjal
Another idea is that it could be a tipping-farmer or spammers who stopped. I made a Dune script (vibed some scripts from @mvr to a new one), and the label 0 accounts are showing the same pattern. Label 2, for example, shows a more gradual decline. Could be that anti-spam measurements are working, but it's also possible that market circumstances (fewer tips, lower prices), makes it less affordable.
1 reply
1 recast
4 reactions
Njal
@cryptonjal
In the 39 days before March 21st on average 58.8% of the casts were by spammy accounts, in the 39 days after (until yesterday) it was on average 40.4%. (I didn't check the first decline because of lack of time).
3 replies
1 recast
5 reactions
Njal
@cryptonjal
Last I checked, the correlation between replies, likes and recasts per spam label, and degen price. In general, label 2 (not a spammer) has the highest correlation, and no label the lowest followed by level 0.
1 reply
0 recast
4 reactions
mvr 🐹
@mvr
What makes it difficult is we don't really know everyones labels on certain dates when the moves happened. Unless one has a spam label last updated in Jan last year. Were comparing to the labels as they are today
1 reply
0 recast
2 reactions