Back in May 2013, the day-ahead read on an FOMC minutes release meant one thing: you sat with the printed statement from three weeks earlier, a legal pad, and a wire feed, and you waited to see whether the word "tapering" had been said out loud in the room. When it turned up in the minutes on May 22, ten-year yields moved forty basis points in six weeks and every emerging-market currency in the book reprice. That was one word, in one paragraph, buried on page nine. Tomorrow's minutes will not be that clean — but the reading discipline is the same one. Let me walk you through the three questions I ask before I open the PDF.
Question 1: Are You Reading the Minutes for a Directional Trade, or a Positioning Check?
This is the fork most retail readers skip, and it is the one that decides whether the rest of your afternoon is honest work or expensive theater. The two purposes are not compatible. If you conflate them, you will trade a positioning check as if it were directional and get run over inside the first ninety seconds of the release.
A directional trade means you believe the minutes will contain new information that moves the front-end curve — a hawkish surprise, a dovish concession, a shift in the language around inflation risk. A positioning check means you already have a view from the last statement and the last press conference, and you are opening the PDF to see whether the room agreed with you or whether the dissent block was larger than the wire summary suggested.
I will concede the strongest argument on the directional side: minutes have moved markets before, and 2013 is the exhibit. That is real. What that argument leaves out is the ten years of minutes since where the release did nothing beyond producing two green candles and one red candle in the first twenty minutes, followed by mean reversion into the London close. Once you correct for the survivor-selected examples the trading-desk anecdotes lean on, the base rate is unkind.
If Yes — You Are Trading the Release Directionally
Reduce your intended size by roughly two-thirds and pre-place your orders. The window in which the release is genuinely tradable is the two minutes before the wire houses finish parsing the PDF and the automated tagging bots overwrite the tape. After that you are trading the second-order interpretation of what other humans decided the minutes meant, and your edge on that is zero.
If you are on a broker with variable spreads — Exness on a standard account, HF Markets on the retail book — expect the EUR/USD spread to widen from the typical one pip to something in the three-to-five-pip range for about ninety seconds. Your pre-placed stop needs to be outside that band or you will get filled on nothing more than the market-maker's own uncertainty.
If No — You Are Positioning-Checking
Close the trading platform. Open the PDF. Read it top-to-bottom without a chart in the room. Your job for the next forty minutes is not to trade — it is to update your prior on what the median FOMC participant thinks about the reaction function. Whether you act on that update happens tomorrow or the day after, not in the release window.
Most of my worst releases have come from starting the day intending to positioning-check and drifting into a directional trade by minute six because the tape looked juicy. Write the intention down on paper before the release. Read it back at minute five.
Question 2: Do You Actually Know What Was Priced In Going Into the Meeting?
This is the question that separates the traders who make money on Fed events from the ones who lose money slowly enough to blame it on variance. Every minutes release trades against a specific expectation set that was already in the curve on the day of the meeting. If you cannot state that expectation set in numbers before you read the minutes, you are not reading them — you are reacting to them.
The math is not optional here. Let me walk you through it with a hypothetical that mirrors how the calculation actually works, so you can reproduce it for the real numbers tomorrow.
Assume the meeting closed with fed funds futures pricing a terminal rate of 4.25% and implying two cuts over the next twelve months. That gives you a starting curve position. The minutes are three weeks stale. In those three weeks, if the two-year yield rose by 15 basis points, roughly 10 basis points of that came from repricing of the terminal rate expectation and about 5 basis points from term premium — that is the rough decomposition the New York Fed's own ACM model would produce on a normal week. So the minutes need to justify a terminal rate about 10 basis points higher than what was in the curve at the meeting itself. If the language is neutral relative to the statement, the front end sells off — because the market already moved and the minutes did not validate the move. If the language is meaningfully more hawkish, the front end holds. If it is more dovish, you get a snap rally of roughly 8 to 12 basis points in the two-year in the first forty minutes.
That is the arithmetic. Five numbers derived from each other: terminal rate, cuts implied, two-year move, decomposition into rate expectation versus term premium, gap to justify. You can do the same exercise on your side of the desk in ten minutes with any Fed funds curve tool. If you have not done it before you open the PDF, you are guessing.
If Yes — You Have the Priced-In Baseline
You are in a position to trade the delta. Write the two-year yield level from the meeting day on paper. Write what it is now. Compute the gap. Read the minutes with the gap as your lens: is the language consistent with the market having moved that much, or does the room read as more cautious than the curve implies?
If No — You Do Not Have a Baseline
Do not trade the release. This is the streetwise part: I know the Telegram groups and the fintwit hot takes are telling you that the release itself is the signal. It is not. The signal is the release relative to a specific priced-in expectation. Without the second half of that sentence, you are trading a rumor.
Spend the release building the baseline instead. Pull the fed funds futures curve as of the meeting close, mark the terminal rate, mark the twelve-month cut count, and file it. You will be ready for the next release, and the one after that. The compounding is real; the shortcut is not.
Question 3: Are You Tracking the Dot-Plot Language or the Balance-Sheet Language?
The minutes contain two conversations happening in parallel, and they do not always agree. One is about the path of the policy rate — the language that maps to the dot plot and to fed funds pricing. The other is about the balance sheet: the pace of runoff, the composition of holdings, the discussion of when to end quantitative tightening or shift the maturity mix. Retail readers almost always read only the first conversation. The second one moves the long end and the dollar in ways the first one does not.
The reason this matters for tomorrow specifically is that the two conversations decouple during turning points. Rate-path language tends to lead the balance-sheet language by one to two meetings. If the room is starting to think about ending runoff, that discussion shows up in the minutes before it shows up in the statement or the press conference. That is the highest-signal paragraph in the document, and it is usually two pages after the section that gets tagged by the wire services.
If Yes — You Are Reading for Dot-Plot Signal
Focus on the "participants" versus "members" distinction and on the size of the dissent block. The word count on hawkish concerns versus dovish concerns is a rough but useful proxy. If the participants section describes hawkish concerns in three paragraphs and dovish concerns in one, the median has shifted regardless of what the summary line says. The front-end curve is your instrument here — trade two-year yields, or if you are in FX, trade the dollar against a low-yielder like JPY or CHF.
If No — You Are Reading for Balance-Sheet Signal
You need a different toolkit. Search the PDF for "reserve balances," "runoff," "SOMA," and "reinvestment." Any discussion of adjusting the cap on Treasury runoff, changing the MBS reinvestment mechanics, or the level of ample reserves is a long-end story. The instrument is the ten-year, the thirty-year, and the dollar against high-yielders — EUR, GBP, AUD.
I have watched traders spend an entire release cycle mining the dot-plot section for a story and miss a two-line paragraph on runoff caps that moved the ten-year by 8 basis points intraday. The signal is where you look. Look in both places, but know which one you are trading before you start.
If You Answered Everything: The Response Map
| Q1: Directional or Positioning? | Q2: Priced-In Baseline? | Q3: Dot-Plot or Balance-Sheet? | Recommendation |
|---|---|---|---|
| Directional | Yes | Dot-Plot | Trade the two-year in the first two minutes; pre-place stops outside the widened spread. |
| Directional | Yes | Balance-Sheet | Trade the ten-year or dollar against a high-yielder; watch the SOMA paragraph specifically. |
| Directional | No | Dot-Plot | Do not trade. Build the fed funds baseline during the release and be ready next cycle. |
| Directional | No | Balance-Sheet | Do not trade. You cannot read runoff signal without a term-premium starting point. |
| Positioning | Yes | Dot-Plot | Read top-to-bottom; update your prior on the reaction function; act tomorrow, not today. |
| Positioning | Yes | Balance-Sheet | Same discipline; the balance-sheet update usually informs a slower, multi-day position. |
| Positioning | No | Dot-Plot | Read the minutes as a document, not as a trade. Build the baseline for next time. |
| Positioning | No | Balance-Sheet | Same as above. Long-end positioning without a term-premium baseline is guessing. |
One paragraph of context on the table: the most common failure mode in this matrix is the top row. Traders answer "directional / yes / dot-plot" because it sounds like the sophisticated answer, then treat the priced-in baseline as an intuition rather than a written number. The discipline is not knowing the framework — it is writing the numbers down before the release and refusing to trade if the paper is blank.
The calendar ahead — three dated events that will test this reading:
- Next FOMC statement release, three weeks from tomorrow's minutes. Watch whether the language shift you tag in tomorrow's minutes shows up in the statement itself. If it does, your read was priced in early. If it does not, either the room walked it back or the wire houses over-read the paragraph.
- The next Summary of Economic Projections release, next quarter's meeting. The dot plot is where the participants-versus-members distinction gets resolved into numbers. Your reading of tomorrow's dissent block will either be confirmed or falsified.
- The next BIS quarterly review, in roughly ten weeks. The BIS reads the Fed the way this desk reads the BIS — as a primary source, not a summary. Their commentary on the balance-sheet trajectory is the outside check on whether the runoff paragraph in tomorrow's minutes was the signal or the noise.
FAQ
How long after the release do FOMC minutes actually move markets?
The tradable window is roughly the first two to twenty minutes. Wire services tag the document in under sixty seconds; automated systems finish parsing in about two minutes; after that the market is trading the second-order interpretation rather than the text itself. Sustained moves — the kind that persist into the next session — usually reflect a repricing of the terminal rate or the balance-sheet trajectory, not the initial spike. If nothing has held after forty minutes, nothing will.
Why does the 2013 taper example keep coming up in this context?
Because it is the cleanest case in the modern archive of a single word in the minutes moving the entire global curve. The May 22, 2013 minutes surfaced the word "tapering" and produced a roughly 40 basis point move in the ten-year over the following six weeks. That release is the archetype every desk uses to teach juniors what a genuine signal in the minutes looks like. Most releases since have not matched that magnitude, which is precisely why the base-rate discipline in Question 1 matters.
What is the difference between the "participants" and "members" language in the minutes?
"Members" refers to the voting members of the FOMC in that meeting; "participants" refers to the broader group of Reserve Bank presidents who attend and speak. The distinction matters because "participants" language captures the fuller distribution of views on the committee, including non-voting hawks and doves whose positions will influence future meetings. A shift in the participants section that has not yet made it into the members' actions is often the earliest lead indicator in the document.
Which currency pairs actually respond to FOMC minutes releases?
The cleanest response is in the dollar against low-yielders — USD/JPY and USD/CHF — because those pairs isolate the front-end rate story with the least noise from the other central bank. USD/EUR responds but with more contamination from ECB expectations. USD/AUD and USD/CAD respond more to the balance-sheet paragraph and to commodity-linked flows. Emerging-market pairs move on the aggregate risk signal rather than the specific language, which is why the 2013 taper hit them so hard.
Should I use a market-maker broker or an ECN broker for trading Fed releases?
For the first two minutes of a release, the spread widening on retail market-maker books is meaningful — you can see EUR/USD spreads move from one pip to three or five pips on brokers like Exness or HF Markets during the release window. ECN routes through IC Markets or Pepperstone tend to widen less but still see slippage on stop orders. If you are trading the release directionally, pre-placed limit orders outside the widened band are the discipline; market orders during the first ninety seconds are how retail accounts end the day negative.
What is the balance-sheet paragraph and where do I find it in the PDF?
It is usually in the discussion of "System Open Market Account holdings" or under a heading referencing reserve balances, runoff caps, or reinvestment. Physically it tends to appear in the middle third of the document, after the economic outlook section and before the policy actions section. Wire services often summarize it in a single line at the bottom of the release blurb, which is why the retail reader misses it. The signal is in the paragraph, not in the summary.
Is it worth trading the minutes at all as a retail participant?
Honestly, for most retail readers, no — the positioning-check use of the release compounds better than the directional trade. Using minutes to update your view of the reaction function over months of releases builds a durable read on the Fed. Trying to catch the two-minute directional move against professional flow that has faster pipes and better pre-priced baselines is negative expected value for most desks. The framework in this piece is more useful as a reading discipline than as a signal service.