Every quarter, the Federal Open Market Committee releases its Summary of Economic Projections, and the financial press immediately fixates on a single page: the dot plot. Each dot represents one FOMC participant's view of where the federal funds rate should be at the end of each of the next three years and in the long run. The chart looks simple. It is read badly. The most common mistake is treating it as a forecast of what the Fed will do, when it is actually closer to an opinion poll about what individual members think they should do, conditional on their individual economic forecasts. Those are different statements, and the gap between them is where most of the trading errors happen.
The dot plot is not a commitment. It is a snapshot of nineteen people's views on a date when they may not have spoken to each other for six weeks. By the time you are reading the plot, two or three of those views have probably already changed in response to data that came out the day before the meeting.
What the Dots Actually Are
There are nineteen dots per year — one for each of the seven Federal Reserve Board governors and twelve regional Reserve Bank presidents. Importantly, only twelve of those nineteen vote in any given year (the Board, the New York Fed president, and a rotating cast of four regional presidents). All nineteen contribute dots, voters and non-voters alike. The plot does not tell you which dots belong to voters, which is the first source of confusion: the median dot can shift because of a non-voter changing their mind, even though the actual policy decision is made by the voters alone.
Each participant submits a dot reflecting their view of the appropriate policy rate at year-end, conditional on their individual outlook for growth, inflation, and the labour market. The participants do not coordinate. Two members can submit the same dot for very different reasons — one because they expect strong growth and the other because they expect persistent inflation. The plot collapses both stories into the same point.
How the Median Moves
The headline number from any dot plot is the median. The median is what gets reported, what the bond market reacts to, and what the press calls "the Fed's forecast." It is also a deeply lossy summary of the underlying distribution.
A median dot can move for three structurally different reasons. First, the mass of opinion can shift broadly in one direction — most members revise their view by 25 bps and the median moves by 25 bps. This is the "real" move, the one that matters for forward guidance. Second, a single member at the boundary can move and shift the median by an outsized amount — if the median was 4.50% and the member just above it drops to 4.25%, the median falls by 25 bps even though no one else changed their mind. Third, a member who was previously below the median can simply not submit a dot in the same range — composition changes due to rotation or attendance can shift the median without anyone changing their view.
Median dot vs realised rate
end-of-year, fed funds %The lesson is that you cannot read the median in isolation. You have to look at the dispersion — how tightly clustered the dots are around the median — and the tails. A 25-bp drop in the median accompanied by a tightening of the dispersion is a high-confidence signal that the committee is converging. The same drop accompanied by widening dispersion is the opposite — it is the median moving while the underlying disagreement is getting worse.
The Long-Run Dot
Buried at the right edge of the plot is the long-run dot, which is supposed to represent the policy rate consistent with full employment and stable inflation in steady state. This is the "neutral rate," r-star, the most theoretically interesting number on the page. It also tends to be the one that moves the slowest and reveals the most.
For nearly a decade after the GFC, the long-run dot drifted lower and lower, eventually settling around 2.5%. That drift was the FOMC slowly conceding that secular forces — demographics, productivity, the global savings glut — had pushed neutral well below pre-2008 levels. Then in 2024-2025, the long-run dot started to drift back up again, suggesting that members are revising their view of structural neutral in light of post-pandemic dynamics: persistent fiscal deficits, deglobalisation, energy investment cycles.
Watching the long-run dot is the closest the FOMC will come to telling you whether they think the post-2008 world is over. Right now they are saying maybe.
Dispersion Is the Real Signal
The most under-utilised piece of information in the SEP is the dispersion of the dots. A tight cluster around the median means the committee is in agreement. A wide spread means there are competing narratives inside the building. The market typically prices the median; the actual policy outcome over the next year is much more sensitive to which narrative ends up winning.
Dispersion of 2026 dots
bps spread between high and lowWhen dispersion is widening, the committee is fragmenting. That is a leading indicator of a regime change in policy — the median has not moved yet, but the centre of gravity is starting to. When dispersion is narrowing, the committee is converging. That is a leading indicator of decisive action: cuts or hikes are coming.
A sharp move in the median with no accompanying narrowing of the dispersion is the worst kind of signal. It means the median moved for technical reasons, and the actual policy direction is unclear. Pricing aggressively off that kind of move is a good way to lose money.
How to Read a New Dot Plot
When a fresh SEP comes out, work through the page in this order. First, look at the change in the long-run dot. If it moved at all, that is the headline — it is rare and meaningful. Second, look at the dispersion of the year-end dots, especially for the next two calendar years. Are they tighter or wider than last quarter? Third, look at the median for each year — but explicitly compare it to the dispersion you just measured. Fourth, look at the tails. Are there outliers far above or below the median? Outliers are often the leading indicators of regime change, because the people closest to the data are the first to revise.
Only after all of that do you compare the new dots to the market's pricing of fed funds futures. The market is almost always tighter than the dots in the near term and wider in the long run. The interesting trades are at the points where the gap is largest and where the dispersion of the dots suggests the market may be wrong about the central tendency.
The single biggest mistake retail investors make with the dot plot is treating it as binding. The Fed will deviate from the dots whenever the data tells them to. The dots are an honest snapshot of intentions at a moment in time, conditional on a forecast that is itself uncertain. They are not a promise.
What the Dots Cannot Tell You
The dot plot has structural blind spots that get worse the further out you look. It cannot capture conditional reaction functions — "I would cut faster if unemployment rises above 4.5%" — because each dot is a single point estimate. It cannot capture optionality — the value of waiting before moving — because it forces participants to commit to a single number. And it cannot capture the political economy of dissent, where members may submit a dot they do not really believe in because it preserves their ability to shape the next meeting's discussion.
The result is that the plot is most useful for the next year and progressively less useful as you move out. The current-year dots are pretty informative because the data window is short. The next-year dots are weakly informative because they depend on a forecast that is mostly noise. The two-years-out dots are mostly aspirational. The long-run dot is philosophical.
Trading the Plot
The relevant trades around an SEP release are usually in the front end of the curve and in fed funds futures. The most reliable pattern is that when the median moves and the dispersion tightens at the same time, the market under-reacts in the short term and the trade is to fade the under-reaction. When the median moves and the dispersion widens, the market over-reacts in the short term and the trade is to fade the over-reaction.
That is not financial advice, and it is not always true. But it is a useful framework for reading the chart in the way that the people who actually trade off it read it. The Fed is telling you something. It is just not telling you what most people think it is telling you.
Final Note
The dot plot is one of the best examples in modern finance of how a piece of data designed to be helpful can be systematically misread. The Fed introduced it in 2012 to improve transparency. It has had the opposite effect for most retail readers, who treat the median as a forecast and the long-run dot as a target. Treating the plot as what it actually is — a moment-in-time, opinion-poll snapshot of nineteen individual conditional forecasts — is the small habit that separates people who are reading the document from people who are reading the headline.