GRID is tempting to treat as a trading signal. It has four tidy quadrants, familiar economic labels, and a clear story for what ought to work in each one.
The trouble sits in the history. Many underlying series are revised, so today’s chart is not necessarily the chart investors saw at the time. GRID stays context-only until the dashboard can preserve that distinction properly.
Two axes, four blocks on each axis
The growth axis divides evidence into four equal blocks:
- private demand and output;
- household income and spending;
- labor impulse;
- leading cyclical and credit impulse.
The inflation axis also uses four blocks:
- current inflation momentum;
- inflation breadth and persistence;
- labor-cost pressure;
- pipeline pressure and expectations.
Each block contributes 25% of its axis. Missing observations are normalized within a block instead of giving another block extra weight.
That structure prevents a long list of labor series or closely related inflation measures from controlling the whole map.
Observation date is not release date
An economic series may describe March, be published in April, and be revised in May. A backtest that places the final revised March value into a March 31 portfolio decision has used information that was not available.
FRED’s ALFRED documentation explains how archived vintages preserve the period when data were originally released and later revised. Its guide to real-time periods shows how an application can request what was known on a past date.
The dashboard publishes release and observation timing where possible, but it does not yet store a complete point-in-time vintage history for every GRID input.
Why revisions are unavoidable
GDP is revised as the Bureau of Economic Analysis receives better source data. The BEA’s GDP page publishes advance, second, and third estimates before later annual revisions.
Inflation data also change through seasonal adjustment and methodology updates. The Bureau of Labor Statistics CPI program documents its releases and revisions.
Quarterly lending surveys create another timing issue. The Federal Reserve’s Senior Loan Officer Opinion Survey describes a survey period, release date, and quarterly result. Carrying the value backward to the reference quarter would make the signal look earlier than it was.
What GRID is good for
GRID gives the current market regime an economic cross-check. Reflation can explain why cyclicals, commodities, and nominal growth are strengthening. Inflation can explain why margins and duration are under pressure. Deflation can explain demand for cash and high-quality bonds.
The map can disagree with market prices, and that disagreement is often the interesting part. Markets may turn before the data. Economic data may improve while financial conditions tighten. Neither side automatically wins.
Why it cannot change KISS yet
The KISS history is intended to represent signals available at each historical close. Revised macro history cannot meet that standard without stored vintage data and release calendars.
Allowing GRID to affect the allocation now would create a backtest that knows more about the past than investors did. The dashboard chooses the less exciting answer: keep GRID visible and keep it out of the allocation rule.
Use GRID for the job it can do
Read GRID as economic context. Ask whether growth and inflation agree with market behavior.
Do not turn the quadrant into a standalone trade signal. Giving it an allocation role would require archived vintages, release-date handling, and a new methodology version.