Michael Sullivan

I didn’t want to write this. One of the deepest lessons I’ve learned from researching sentiment data is how irrational and angry people get towards the bottom of Bitcoin bear markets. People that are historically aligned often fall into conflicts that wouldn’t exist near tops. The BIP-110 debate increasingly feels like one of those moments. I wanted no part of that. I’ve been publicly cancelled before, and I have no desire to go through that again. I know how painful it is to have an angry crowd criticizing and shaming you, which gives me a strong bias to avoid situations exactly like this one. But I’m pushing against that bias because (for better or worse) I have a novel dataset that offers a unique lens on the BIP-110 debate. I’m going to attempt to keep this unbiased, but before I do, I’m going to lead with what will be my highest-conviction opinion of the entire piece: People like Mechanic and Luke Dashjr understand Bitcoin’s consensus mechanisms better than I do, while Adam Back and Jameson Lopp grasp Bitcoin’s technical nuances more deeply than I ever will. I am not here to pretend I understand these dynamics of this debate better than the people who have spent years studying them. That’s not why I’m writing this. “Then why should I care what you have to say?” Because there is another side of the debate that is also important, and it’s on this side I have a different perspective. The social layer. For those of you new to my work, I’ve been researching the language, narratives, and emotional patterns of different groups throughout Bitcoin. I’ve been (obsessively) studying the emotional regimes and crowd psychology associated with the space. That includes the people now publicly associated with both sides of the BIP-110 debate… and the differences between these groups are striking. I believe I would be doing a disservice by keeping that data to myself simply because publishing it might piss people off. So there will be no paywall on this piece, no call to action halfway through, and I won’t ask you to become a subscriber. I simply want these charts to exist publicly, where anyone can examine them or interpret them differently than I do. I will try to present it as fairly as I can, but I am not neutral, and neither is anyone reading this. People will inevitably see different things in these charts. This data cannot tell us who is technically correct. But it can help us to surface clues about how the debate is evolving, who thinks they are winning, and why I believe one of these groups is currently experiencing an engagement-reinforced narrative environment.

Sentiment is lower now then during the bottom in 2022. One of my favorite ways to visualize this is by looking at the rate of "optimism" in Bitcoin OGs language over time... but what is more striking is the engagement rate. Exact same users. Deteriorating engagement.

No worries at all! Great questions. The whole process is a bit complex and I cover it in great detail in a couple of those podcasts linked in the piece if you wanna get my exact methodology (I may need to get a reference video or some thing now that I think of it). I don't share the raw source data for a few reasons, one of which being that it's extremely expensive and not all open source and I'm just trying to reduce my own risks. The cohorting thing is there I spend a TON of time. I've done this for all of the groups I look at, but basically I do an analysis (assisted with tons of AI compute) to validate that there are not individuals in the cohort that don't accurately represent the larger group. If these get flagged I then manually go through and analyze them myself. So people that throw BIP-110 in their bio as a joke are definitely not in there.