Why there is no consensus column

Why it matters

Open the newest CPI row on Kitalpha’s calendar and, where another calendar would print an expected value, it says: "No consensus estimate is shown. Kitalpha does not license survey forecasts of economic releases.". Open the newest FOMC brief and a different kind of figure appears: probabilities quoted from named prediction markets, cited line by line, 4 of them. A reader used to a consensus column asks two questions: where is the expectation, and is the brief’s figure the same thing under another name? The answers are “nowhere, by design” and “no”, and this lesson explains both.

The concept

On most economic calendars, an expectation is a survey statistic. A data vendor polls a panel of forecasters before each release, takes the median of their answers, licenses the result to publishers, and the publishers print it beside the release as the consensus. A surprise on such a calendar is the printed value minus that median. Two things about the arrangement matter here: the figure comes from a licensed survey, and it is a forecast of the number.

Kitalpha does neither part. It licenses no survey of forecasters, so it has no consensus to print, and it publishes no forecast of its own, for any release, on any page. Rather than leave an empty column that a reader might take for missing data, every row states the fact in a sentence. The calendar is a schedule, read from the agencies’ own release calendars, plus one published statistic about the past.

That statistic, the impact tier, is the field most often mistaken for an expectation in disguise, so it is worth saying exactly what it is not. The tier is computed from past release days: the average size of the 10-year Treasury yield’s move on those days, in excess of a quiet day, over a trailing window. It has no sign, because sizes are taken with the sign dropped; it makes no reference to the coming print, because it is computed before the print exists; and its inputs are yields, not the indicator’s own values, so it is not a base rate for the print either. It sits on a scheduled row for months, unchanged, and it would sit there the same if the release were cancelled.

There is one place on the site where a probability appears, and it is a different kind of figure again. A signed brief on a central-bank decision may quote market-implied probabilities: the price at which a contract on an outcome traded on a named prediction market at a named time. The brief names the market, gives the timestamp, and footnotes the line, so the figure is attributed to its source and never adopted as Kitalpha’s view. A market price is not a survey median: it is what traders on one venue paid, it differs between venues, and it moves by the minute. The brief reports it as a snapshot and predicts nothing from it.

So the calendar’s missing column is not a gap waiting to be filled. It is a decision, stated on every row, that the site publishes schedules and statistics about the past, and attributes any probability it quotes to the market that priced it. The table below sets the four things a reader might mistake for an expectation side by side; look for which of them is a probability at all, and where it lives.

Worked example

Take the newest CPI row, whose consensus line reads as above, and the newest FOMC brief dated 2026-09-16. The row carries nothing to compute, which is the point; the steps below read the brief’s market lines instead: how many there are, the first probability quoted, its remainder to one hundred, the first probability from the second market named, and the gap between the two markets. Every figure comes from the brief’s own footnoted lines.

What to read off the steps. The second and third lines are one market’s price for an outcome and for its complement, at one timestamp; they describe that market and nothing else. The fourth line is a second market’s price for a comparable outcome, and the last line is how far the two differed at the moment the brief read them. That gap is the plainest answer to “is this a consensus”: two attributed prices are not one expectation, and the brief adopts neither. The calendar row, meanwhile, carries no probability at all.

Faded example

Now the second market’s line in the same brief, a distinct attributed source: its first quoted probability is given. Complete the last step: the remainder to one hundred.

Retrieval check

Mark your confidence before each answer. Every option carries an explanation; read the ones you rejected too.

Your summary