WET IMPLIEDLive reading · not a governed benchmark

Interest rates and inflation, in the units they are actually quoted in

This desk converts live prediction-market contract prices into two numbers a macro reader can use directly: how much the market expects the Fed to move at each upcoming meeting, in basis points, and what CPI print the market expects next, in percent, with the range that holds 80% of the priced distribution. Both are read off contracts listed on Kalshi, and every figure below links to the contract it came from.

Live from the contracts11 meeting strips · 2 CPI ladders · 08:24:50 UTC · every 12s
Next meeting · Sep 2026
+21.4bps

The average policy-rate change priced for this meeting: +21.4 basis points, i.e. 0.214 of a percentage point of tightening.

hold 12% · cut 1% · hike 87%

KXFEDDECISION-26SEP on Kalshi ↗

Priced by Jan 2028
+50.3bps

The per-meeting expectations added up across all 11 readable meetings — how much policy-rate change is priced in cumulatively by Jan 2028.

Next CPI · Sep 2026
3.54%

The expected year-over-year CPI print, read across 21 strike contracts. 80% of the priced distribution sits between 3.36% and 3.69%.

KXCPIYOY-26SEP on Kalshi ↗

Same contracts, two readings182-day constant maturity · methodology v2.0
WET Implied · live, ungoverned
+47.5bps

The cumulative path above, interpolated to a fixed 182 days out, right now.

Not an intraday WETFED: that benchmark only prints at settlement. This is the WET Implied reading of the same path at the same horizon, and it never enters the ledger or the checksum chain.

WETFED · the governed close
+33.9bps

The WET Close for 2026-09-07 — the same interpolation, printed once and written to the checksum-chained ledger. How that close was taken is stamped on the row itself; see below.

Every FOMC meeting on the board

Kalshi KXFEDDECISION · normalized so the outcomes sum to 100%

Read down the “by then” column to see the term structure: how much of the priced policy change is near-dated versus deferred. Wings quoted as “more than 25bps” are booked at 50bps, the published convention. Volume is contracts traded, not dollars.

Implied CPI, print by print

Kalshi KXCPIYOY · “above X%” strikes, repaired to a distribution
CPI year-over-year · Sep 202621 strikes · 106k contracts
3.54%80% band 3.36–3.69%
above 3%above 5%

Open the ladder on Kalshi ↗

CPI year-over-year · Nov 202621 strikes · 11k contracts
3.49%80% band 3.11–3.88%
above 3%above 5%

Open the ladder on Kalshi ↗

Where these numbers come from

one venue, two contract series
The rate path · KXFEDDECISION
One contract group per FOMC meeting, each covering a mutually exclusive outcome — cut more than 25bps, cut 25bps, hold, hike 25bps, hike more than 25bps. Because exactly one of them must happen, their prices form a probability distribution over the meeting once normalized, and the average of that distribution is an expected rate change in basis points. Open the series on Kalshi ↗
The CPI reading · KXCPIYOY
A ladder of “CPI comes in above X%” contracts. Read together they trace out the priced probability of every level, which gives an expected print and a band rather than a single guess. Open the series on Kalshi ↗
Prices used
Two-sided mids — the midpoint of the best bid and best ask — falling back to the last trade only when the book is empty. A meeting whose prices do not sum to a coherent distribution, or a ladder with fewer than four usable strikes, is dropped and reported as no value rather than estimated.

W.E.T. is not an exchange and holds no funds. Venue links are affiliate-tracked; that never changes what is reported. Machine-readable snapshot ↗

How this relates to WETFED

same contracts, different governance

WETFED is the governed version of the rate path on this page. It is built from the same KXFEDDECISION meeting groups and the same cumulative path — then it does three things this desk deliberately does not:

  • Fixes the horizon. Meetings expire; a benchmark cannot. WETFED interpolates the cumulative path to a constant 182 days out, so today’s value is comparable to last year’s.
  • Prints once, on a rule. One value a day, at the close, written down and never revised in place. How that close was taken is stamped on the row itself rather than described here — see below.
  • Keeps a record. Every close is written to a checksum-chained ledger under a versioned, published methodology, with a restatement policy. That is what makes it quotable.

So: watch here, cite WETFED. The live and settled readings sit side by side in the desk above — WETFED last closed at +33.9bps on 2026-09-07.

How that close was taken

This close is ONE instantaneous observation, not a settlement window. WETFED is a derived constant-maturity ladder over the implied Fed path, not a set of marked contracts; the mark tape is keyed by venue marketRef and carries no observations of a meeting ladder, so a settlement window cannot exist for this benchmark under any tape cadence. A single-read close can be moved by one trade at the final second, and that exposure is published rather than described away.

2026-09-07 · closeBasis: single-observation · closeObservations: 1

A window is unavailable by construction, not by accident — the reason is the one the row gives above. Making one possible is a data-collection change, not a formula change: the same input would have to be sampled more than once inside the close period and stored so both samples survive. That collection does not exist on this desk today, so a single-observation close can still be moved by one trade at the final second.

WETFED, the benchmark →Published methodology →

How the reading is taken

the whole method, in two paragraphs

Meeting groups. The five outcomes for a meeting should price to 100% between them. In practice they sum to a little more or less, because bid-ask spreads and stale quotes do not cancel. We divide each price by the group total so the outcomes sum to exactly 100%, then take the probability-weighted average of the rate changes those outcomes represent. Open-ended wings (“more than 25bps”) are booked at 50bps — a published convention, and the conservative one. A group whose total lands outside 0.5–1.5 is treated as unreadable, not adjusted.

Threshold ladders. “Above X%” contracts should get cheaper as X rises. When quotes cross that ordering we repair the sequence with isotonic regression (pool-adjacent-violators) rather than dropping the offending strike, difference the repaired curve into a distribution, and read the expected value and the 10th-to-90th percentile band from it in the series’ native units. Ladders too hollow to form a distribution return no number.

WET Implied is an indicative reading. It is not a governed benchmark: it does not settle, it is not written to the ledger, and it is not part of the checksum chain. Data, not advice — prediction market trading carries risk of loss, and prices should always be verified at the venue.

Where to go from here

BenchmarkWETFED — the governed Fed pathThe same contracts, settled daily to a fixed horizon, with a track record you can cite.MethodologyHow WETFED is computedFormula, the weight basis, consolidated pricing, continuity, the close basis and the restatement policy.BasketFed Path basketThe real rate-policy markets behind the theme, held as one spread and tracked as one hypothetical line.BasketInflation Watch basketThe CPI and inflation markets this ladder sits inside, with their combined hypothetical record.BasketGlobal Easing basketRate-cut pricing beyond the Fed — the same question asked of other central banks.BoardThe full index boardEvery benchmark and slate the desk publishes, with change, regime and coverage.DataMachine-readable snapshotThe exact payload this page renders, cached on a short window and free to quote with attribution.SupersededThe old Fed Path gaugeThe 0–100 slate this desk replaced. Still published for continuity — read it as history.

Basket records are hypothetical · mid marks · before fees, spread and slippage. They are percentage readings on an equal-notional, size-blind construction, not performance claims and never a dollar figure.

Common questions

What does WET Implied actually measure?
Two things, both in their own real units. First, the policy-rate change priced for each upcoming FOMC meeting, in basis points — the probability-weighted average of the cut, hold and hike contracts for that meeting. Second, the expected year-over-year CPI print, in percent, read across the whole ladder of "above X%" contracts, together with the range that holds 80% of the priced distribution.
Where do the numbers come from?
From live contract prices on Kalshi, a CFTC-regulated exchange. The rate path is read from the KXFEDDECISION meeting series and the CPI reading from the KXCPIYOY ladder. Every meeting and every ladder on the page links straight to the contract it was read from, so any figure can be checked at the source.
How is WET Implied different from WETFED?
They read the same contracts. WET Implied is the live, ungoverned reading, refreshed continuously through the trading day and never written down. WETFED is the governed benchmark: the same cumulative path interpolated to a fixed 182-day horizon, printed once a day as the WET Close and written to a checksum-chained ledger with a published, versioned methodology. The row for 2026-09-07 states its own basis — closeBasis "single-observation", closeObservations 1 — in these words: "This close is ONE instantaneous observation, not a settlement window. WETFED is a derived constant-maturity ladder over the implied Fed path, not a set of marked contracts; the mark tape is keyed by venue marketRef and carries no observations of a meeting ladder, so a settlement window cannot exist for this benchmark under any tape cadence. A single-read close can be moved by one trade at the final second, and that exposure is published rather than described away." Use WET Implied to watch, and WETFED to quote or cite.
Can I trade WET Implied?
No — it is a reading, not a market. W.E.T. is not an exchange and never handles orders. The underlying contracts are listed on the venue, and every reading on this page links to them. The index desk also publishes curated baskets that hold the same themes, each with its own tracked, hypothetical performance record.
What happens when a reading cannot be taken?
The desk publishes the absence. A meeting strip whose prices do not sum to a coherent distribution, or a CPI ladder with fewer than four usable strikes, produces no number at all rather than a guess. If a refresh fails, the last good reading stays on screen labelled with its real age — nothing is carried forward as though it were current, and no gap is ever interpolated.