September 2026 CPI Release

The August print landed one tick above the chain—again. BLS released August CPI this morning: headline 3.4% year-over-year, 0.4% month-over-month, core 2.4% YoY and 0.3% MoM. Kalshi's cumulative chains went into the release centred on 3.3% headline YoY, 0.3% MoM and 2.3% core YoY. All three printed exactly one tick high.

That is now five consecutive upside misses this workspace has tracked across the July and August releases, on four separate chains. A single miss is noise. Five in the same direction is a tilt in how the cumulative book prices the tail above its mode, and it is the reason to watch the October 14 release rather than assume the chain has recalibrated.

The workspace has rolled to the September release. Every chain below now points at the September data, published October 14 at 8:30am ET. The September book is a month out and thin—the headline YoY chain shows a 73% bid above 3.4% and 42% above 3.5%, which puts the modal outcome at 3.5%, a tick above what August just delivered. Read the spreads, not the last trade: several strikes still quote 0.30 wide.

Kalshi's cumulative September chains, displayed in their native form and paired with headline year-over-year. Cumulative markets price "above X", so each step down the ladder is the probability the print clears that threshold—the gap between two adjacent strikes is the odds of landing exactly in between.

A month out these are wide by construction. Headline YoY runs from 97% above 3.0% down to 22% above 3.6%; core YoY runs 66% above 2.1% to 13% above 2.4%. Compare each to where August settled and the September book is pricing inflation to hold or firm, not to cool.

The Categorical books sit beside the core chains. They express the same question as a discrete distribution rather than a ladder, and they were the more accurate of the two forms on the July release—the chain priced at-or-above 0.1% MoM at 15.5% while the Categorical put 39.5% on exactly 0.1%.

On August both forms missed in the same direction, so the July divergence has not repeated. Worth watching whether the Categorical continues to carry a fatter middle than the cumulative ladder implies as September volume builds; right now KXECONSTATCPIYOY-26SEP has turned $3,961 in 24 hours against near-zero on the month-over-month sibling.

The annual context widgets round out the stack: Polymarket's 2026 inflation-ceiling book, which trades the whole-year path rather than any single print, and the Manifold binary on whether CPI comes in higher than the prior month.

• Five upside misses and counting. July headline YoY, headline MoM, core YoY, core MoM and now August headline all printed at or above the chain's modal strike. If the September chain again centres a tick below the eventual print, the pattern is worth trading rather than noting.
• The September chain already prices a firmer number. Modal headline YoY sits at 3.5% against August's 3.4%, and the MoM book keeps a 46% bid above 0.4%. Gasoline did the work in August; whether that persists is the single biggest swing factor into October 14.
• Core is the quieter story. Core YoY at 2.4% is the softest reading on this page in months, and the September core chain centres on the same 2.4%. A core print that holds while headline firms is the shape that makes a rate decision hardest to call.
• Liquidity is thin a month out. Several September strikes quote 0.30 wide, and a few last prices sit outside the current bid-ask. Trust the bid side of the cumulative ladder over last-trade marks until volume arrives in the final fortnight.
• Next release: October 14, 8:30am ET, covering September data. The countdown at the top of the page tracks it.