← Explore the archive

Archive / 2008 · 07

When GodIndex and the US Index moved in opposite directions

A short July 2008 note links opposing index movements to new participation from Asia. It is an early clue that audience composition and weighting mattered to GodIndex.

New editorial reading of a historical post. Original topic and URL preserved; commentary below is not an unaltered reproduction.

The original explanation

The 14 July post attributed a negative relationship between GodIndex and the US Index to increased participation from China, Japan and South Korea. It also described rebalancing. This documents the authors’ explanation and names additional countries of participation; it does not supply counts, a correlation coefficient or a reproducible weighting formula.

The post’s linked chart was hosted under a Globrate address. Because the underlying series and chart provenance are not fully established, we do not relabel that image as a verified GodIndex dataset.

Opposing directions are not a paradox

A component and a composite need not move together. In a hypothetical two-group index, one group’s response can decline while a sufficiently large rise in the other group lifts the total. If the weights change as well, the aggregate can shift even without the same change occurring within either group.

That is why an overall line should be accompanied by its composition and calculation rules. Otherwise a reader may mistake an audience change for a change in the opinions of existing participants.

What can be recovered from a short note?

The article is evidence that the early system was being discussed as an adjustable composite, rather than as a permanently fixed simple average. Its wording does not tell us exactly which mathematical operation was applied, when it took effect or how earlier values were treated.

A modern analysis should preserve those unknowns. It should not calculate a precise correlation from a hand-drawn trend or claim that a named country caused the movement solely because the original author said so. Our methodology page separates known response rules from the missing calculation details.

Sources & verification

  1. Original historical source

Editorial review: 6 September 2026. Found an error or another archive? Please contact us.

The next chapter

The archive is open. Voting returns soon.

Join the launch newsletter, propose a research project, or help support an independent GodIndex.