Reading the Whole Ladder: Why Estrata Publishes Low Scores

Key takeaways

• A credible scoring system produces a distribution, not a highlight reel. Until this week, every opportunity Estrata surfaced scored 80+. That was selection bias, we covered the flagship national programs first, and we are correcting it.
• Seven new records span the ladder: Metro Manila office recovery (60, watchlist), LNG buildout beyond fuel-switching (57), the nuclear/SMR program (46), Bataan plant rehabilitation (33), Manila Bay reclamation restart (22), flagged-tier flood-control contracting (13), and offshore-gaming revival (9).
• Low scores are evidence-backed, not vibes. The 13 on flood-control contracting rests on COA's 421 documented ghost projects, the ICI investigation, and frozen 2026 appropriations¹, the same provenance standard as our 88s.
• The same rubric, the same math. Every score is the weighted sum of six factors (capital scale 25%, government backing 20%, readiness 20%, second-order effects 15%, evidence 10%, competition 10%) under locked version 2026.1, with a published sensitivity range.

What a low score means: and what it doesn't

A score is not a buy/sell call (nothing we publish is). It is a measure of how much substantiated, executable opportunity the evidence supports today. Read the bands as postures:

The bottom band does double duty. Some entries are risk verdicts: flood-control contracting scores 13 because the documented record (ghost projects, AMLC freezes, a halted budget line) is a catalogue of reasons capital should not be there.¹ Some are prematurity verdicts: Bataan rehabilitation scores 33 not because the plant is corrupt but because a never-fueled 1984 asset with an unfinished feasibility study has effectively zero execution readiness.² And some are policy verdicts: Manila Bay reclamation scores 22 because a standing presidential halt order is the controlling fact, strong enough that it is adding hundreds of millions of dollars to the New Manila International Airport's land program next door.³

Our read: the most useful thing a low score does is redirect. The 9 on offshore-gaming revival and the 60 on prime-office recovery are the same buildings viewed through two theses⁴, the score gap tells you which thesis the evidence supports. That is the product: not "stay away" as a slogan, but a reproducible accounting of why, factor by factor, with the math shown.

Why the top was crowded until now

Mechanically, our first-covered opportunities (grid backbone, solar-storage, connectivity programs) are the flagship projects of the decade: enormous capital scale, explicit government backing, visible construction. The rubric scores that profile high because it is high. The bias was in coverage selection, not arithmetic. As the tracked set widens toward provincial, sectoral, and contrarian theses, the distribution will keep filling in, and scores will move. A published score is a snapshot with a sensitivity range, not a verdict for eternity; the quarterly recompute is where new evidence cashes in.

What we will not do is manufacture spread. If next quarter's evidence pushes three more opportunities above 80, they will print above 80. The discipline is the method, not the histogram.

What to watch

• Band migrations: the LNG record (57) jumps if greenfield plants reach financial close; nuclear (46) moves on regulator legislation and vendor selection.²
• The flood-control floor: ICI case filings that reach principals would, paradoxically, raise future procurement scores by repricing integrity risk across the sector.¹
• Conversion evidence in the office market: every Bay Area building that flips to education or healthcare strengthens the recovery record and buries the revival one.⁴