The Fed Market Just Flipped ~30 Points on One CPI Print. The Forgotten ~20¢ "No Change" Outcome Is the Interesting Number

In about 48 hours, the Federal Reserve explains whether it just raised rates for the first time since 2023. Polymarket now prices a 25 bps increase at 79%, leaving roughly 20 cents on "no change." At that price, the hold is not a percentage: a $100 stake buys about 500 shares, and returns about $500 gross — $400 of profit — if the Fed stands pat, or zero if it does not.

The remarkable part is how fast the crowd got here. Prices rose dramatically from a 50-60% range one week prior; after Friday's inflation report Polymarket repriced to roughly 83% on the hike. But the inflation report that moved the board matched headline expectations — the overshoot was a single tenth of a point in one monthly core measure. That is a thin foundation for a 30-point swing on the first hike in three years, and it is why the cheap "hold" side deserves a second look before the announcement lands Wednesday at 2 p.m. ET.

What blew the market up

The mechanics behind the favorite are real. The Fed now sits at a 3.50%-3.75% target range, and inflation has re-accelerated: headline CPI held at 3.4% year over year in August while core monthly inflation rose 0.3%, one-tenth hotter than the 0.2% consensus. Energy is doing the heavy lifting — oil reclaimed $100 a barrel as the Middle East conflict tightened supply, pushing gasoline up more than 27% from a year ago — and that pressure is broad enough to worry policymakers who target 2%.

The personnel tilted the same way. New Fed chair Kevin Warsh, who took over for Jerome Powell earlier this year, used his August 28 debut at Jackson Hole to say there is still "work to do" on inflation, explicitly opening the door to hikes. Three of the twelve already dissented in favor of a raise in July in a 9-3 vote — and major investment banks including Goldman Sachs and J.P. Morgan have shifted their forecasts to expect the move. Fed funds futures run even hotter than Polymarket.

Put that all together and a hike is the honest base case. The club's chair is a hawk who just campaigned on rate hikes, the last inflation print beat on the margin, and the hawks were already voting that way a month and a half ago.

The boring facts the crowd is sprinting past

Here is the tension. The committee's revealed preference seven weeks ago — a 9–3 vote for patience — does not dissolve because one monthly core reading came in a tenth hot. The headline number the report was judged on matched expectations and even held steady against July. And the seat-of-government math is steeper than a tenth of a point conveys: the Fed has not raised rates since 2023, and in the June projections only nine of eighteen officials saw the funds rate ending this year above its current range.

That last one matters for the "no change" side. If roughly half of the committee's own members did not anticipate a single hike for all of 2026, then a September hike is not a done deal at ~80 cents — it is a bet that one overshoot on one sub-index plus a hawkish speech overrode a majority that voted patience within the last two months. The dissenting hawks could get their way. But the market is paying them like the outcome is nearly settled.

There is also a rule worth reading before picking a side. The contract settles on the change in the upper bound of the target range, using the FOMC statement released after the September 15-16 meeting — rounded up to the nearest 25 basis points, and resolving to "no change" if no statement arrives before the next meeting. No ambiguity, no exotic trigger. A 25-basis-point hike pays the "25 bps increase" bucket; anything else pays the hold.

The money, both ways

The asymmetry concentrates on the forgotten side. At ~20 cents, buying "no change" costs $100 for about 500 shares; if the Fed calls off the hike, you collect about $500 gross against a $100 stake — roughly a 4x turn on the money held for two days. The downside is equally clean: if Warsh and the hawks push the hike through, those shares settle at zero and the stake is gone.

The favorite is the lower-multiple trade. At ~80 cents, $100 on the hike buys about 125 shares and earns roughly $25 if it hits — a ~25% edge the crowd is already charging you for, with the same total-loss risk if the committee blinks and holds.

That is the honest split: the ~80-cent hike is the consensus with a thin profit, and the ~20-cent hold is the contrarian with real multiple but a real chance of arriving at the wrong answer. The list of reasons the market is right — a hawkish chair, a hot core print, three sitting dissenters, aligned banks — is genuinely long. The reason it could be wrong is simpler: the crowd repriced thirty points on roughly one-tenth of one percentage point of a single monthly measure, against a committee that voted 9–3 for patience and has not raised in three years.

The board will be wrong about exactly one of those buckets on Wednesday at 2 p.m. ET, when the statement drops and the other side of whatever price you chose goes to one dollar or zero. The clock is the whole game here — and it reads two days and one announcement.