Every piece is written to the same standard so the reasoning — not the conclusion — is what you judge. This page describes that format.
Every piece opens with the same discipline: what is true at the moment of writing, stated plainly and dated. Prices, levels, positioning, and the relevant on-chain or fundamental data — each one sourced, each one timestamped, so the ground we stand on can be checked rather than taken on faith. Fact is kept separate from interpretation. We establish the present before we argue about the future, because a thesis built on a careless reading of the present is worth nothing.
Before we take a side, we argue against ourselves. Each piece sets out the strongest version of the bull case and the strongest version of the bear case — not a token objection, but the argument a serious opponent would actually make. The bear case is given equal weight, and often greater, because the costs are asymmetric: conviction is cheap, capital is not. If the bear case cannot be answered, there is no thesis to publish.
A thesis is a claim that can be wrong. It names a direction, a level or a range, a horizon over which it should play out, and a degree of conviction stated openly — how strongly the view is held, and why not more strongly. Vague optimism is not a thesis: "this looks interesting" commits to nothing and can never be checked. We commit to something specific enough that time alone can settle whether we were right.
Each piece states, before the outcome is known, the condition that would invalidate its thesis.
Invalidation is defined in advance and in the open: a specific, observable condition — a level broken, a horizon passed, a fact reversed — that ends the thesis. It points one way. We do not hedge and then claim to have been right whichever way the market moved; a view that cannot be falsified explains nothing. When the condition is met, the thesis is closed, and the record says so.
Outcomes are recorded as dated entries at the bottom of each piece and surfaced in the track record. Nothing already published is edited or removed.
The timestamp is verifiable, not declared: every piece is announced in the open Telegram channel at the moment of publication, and each piece links back to its original post. Telegram fixes the time on a platform we do not control — check the claim against the post, not against our word.