Methodology
Carried forward is not the same as confirmed.
The Alt Sector Radar is a research board, not an engine. An analyst places each name against its sector peers, every board carries the date it was reviewed, and a name that has not been re-examined is shown as carried forward rather than quietly presented as still standing.
Alt research goes wrong in a familiar way. Attention follows whatever moved hardest that week, last month's conclusions are never revisited, and a name stays on a list long after anyone last looked at it properly. The result feels like research and behaves like a scrapbook.
The Radar exists to impose a structure on that. Names are organised into focused sectors and assessed against the other names doing the same job, because a comparison across the whole market mostly measures which sector is in favour. Judged against its peers, a name has to earn its position rather than inherit it from a rising tide.
The output is a picture of where analyst attention currently sits and why. It is a discipline for keeping research current and honest about its own coverage, and it is worked by a person rather than produced by a machine.
There is no automated scoring engine behind the board, no weights and no algorithmic ranking. Building one is on the roadmap. Until it exists, claiming it would be the simplest kind of lie to tell and the hardest to detect from outside. What sits behind each placement is analyst work against a defined process, described here as exactly that.
Names sit in one of three plain phases describing how much attention each currently warrants. There is no score, no rank and no ordering within a phase. A numbered list implies the difference between fourth and fifth means something, and on judgement-led research it does not.
When a name has not been re-examined since the last board, it is marked as carried forward. It has not been reconfirmed and the board does not pretend otherwise. This is the single most important discipline in the tool, because a stale placement that looks fresh is worse than no placement at all.
Reviews run on a defined cadence of run types, from a light pass through to a deeper periodic rebuild, and each board publishes the date and how much of it was actually covered. The disclosure exists because coverage has run partial in the past, and the fix was to show it rather than to stop mentioning it.
It is only as current as its last review
A board is a snapshot with a date on it. Between reviews the market keeps moving and the board does not. The carried-forward marking exists precisely so this is visible rather than assumed, but it remains the fundamental limitation of any research board maintained by people rather than by a machine running continuously.
Coverage has not always been even
Some sectors have historically received more thorough passes than others. That is the honest reason every board now carries a coverage note. Anyone claiming complete, continuous, automated coverage of an alt universe is describing something that does not exist here.
Peer-relative means exactly that
A name placed highly within its sector is placed highly against the others doing the same job. It carries no claim about that name against the whole market, and no claim that the sector itself is well positioned. A strong name in a weak sector is a real and common outcome.
It issues nothing to act on
No buy signals, no price calls, no targets, no ranking of what to own. The phases describe research attention. Reading a placement as a recommendation is reading in something the board does not say anywhere.
A dated board beats a confident one.
Members see the full board by sector, the current phase for every name, the review date and the coverage behind it.
Discussion