Central bank forecast scorecard · data to

Rate Call Tracker

Every bank and research house publishes a view before a rate decision. This page records who said what, when they said it, and what the Fed, the ECB and the RBI then did in 2026. Each entry links to the report it came from.

What SG Analytics could do with this

A track record line for every macro note that goes to a client

SG Analytics writes macro updates, CIO newsletters and asset strategy reports for asset managers, wealth managers and sell-side teams in the US and Europe. Before a Fed or ECB decision those notes say what the consensus expects. They rarely say whose view has earned the most weight.

The work it removes

  • Collecting each house's preview before a meeting and logging the call by hand.
  • Going back after the decision to check who was right, which usually does not happen at all.
  • Rebuilding the same table for the next client who asks.

What a client gets

  • One line in the pre-meeting note: the consensus, the dissenters, and each dissenter's record at the last turn.
  • An alert when a house with a good record breaks from the pack.
  • A quarterly scorecard a CIO can use to decide which sell-side research to keep paying for.

Where it extends. The scoring code does not care which bank it is scoring. The same file format covers the Bank of England and the Bank of Japan, and the same idea applies to earnings estimates, GDP and inflation prints, and price targets. The Fed and ECB tabs here use the houses SG Analytics' clients already read.

What already exists. Bloomberg ranks economists by forecast accuracy on its terminal, and FocusEconomics, Consensus Economics and LSEG give annual accuracy awards. Those are paywalled or once a year, and they score the final number. None of them, as far as we know, publishes how many days ahead a house moved before a turn, which is the measure that separates houses on this page.

Method and limits

How a call is scored