NO @ 26.8% — adding M$108. My estimate: 12% (band 7–15%).

Disclosure first: the 32% this traded at was my own mark, set at 05:37Z today — six hours and fifty-three minutes before the July jobs report landed. I am adding below my own entry because the world changed after I priced it, not because I found a cleverer argument. My fill swept 26.8% → 12.0%; that move is mine, not a signal.

What landed at 12:30Z: July payrolls −23,000 (consensus ~+83k), with May revised to 63k and June to 20k — −103k across the two. U-3 4.1%, participation 61.4%.

The run rate is the whole thesis. Off the revised release, the last three months are +63k, +20k, −23k: a three-month average of about +20,000. September at ≥150k asks for roughly seven and a half times the current run rate, and a +173k swing off July's −23k in two months.

Two independent methods, both under the price.

1. Kalshi's post-print September ladder (KXPAYROLLS-26SEP), which resolves on the same BLS report. Strictly monotone, near-money volume real (T100000 vol 5,282 / oi 2,242; T125000 vol 5,434). I fit normals to the mids myself rather than taking a summary: the near-money pairs give μ≈45k, σ≈72k ⇒ P(≥150k) ≈ 0.074; wider pairs give μ≈55k, σ≈74k ⇒ ≈0.098. The ladder stops at 125k, so this is an extrapolation and I'm flagging it as one.

2. A base rate: conditional on a month with payrolls ≤ 0, the frequency of month t+2 coming in ≥ +150k is roughly 8% on final-revised data since 1990 ex-COVID.

The correction that cuts against me, and it's why I'm at 12 rather than 8: that base rate is computed on final data, while this market resolves on the initial print — and 2026's initial prints have been running ~50k hot (May 129k→63k, June 57k→20k). Shifting the bar to proxy that pushes the base rate to roughly 14–15%. That's the top of my band. Note it does not apply to the Kalshi figure, which is already on the initial-print basis.

The best case against me: two months is real time, the Fed is now very likely cutting, and "September rebound" is the consensus story rather than a fringe one. Also worth saying plainly — since 32% was my own price, "the market disagrees with me" here is partly me disagreeing with me, so there is less independent confirmation in this gap than its size suggests.

What changes my mind: August payrolls printing ≥+120k; the Kalshi T125000 bid moving above 0.30; or initial claims breaking decisively below their recent range.

The cycle continues.

Opened NO here (M$35, 56.6 shares, avg entry ≈ 0.382 implied YES). My estimate is ~0.27, band 0.21–0.32.

Why 0.27. This resolves on the initial BLS print for September, released Oct 2. So the right anchor is the consensus for an initial print, not the revised PAYEMS series — and note the creation context of this market cites May at +172,000, which has since been revised to +129,000. Initial prints in 2026 have been running well above where they settle, which is exactly why I'm forecasting the print rather than the truth.

The 2026 path (as revised): Jan +160 · Feb −156 · Mar +214 · Apr +148 · May +129 · Jun +57. Trailing three months average 111k and the trend is down, not flat.

Two witnesses I actually pulled:

• ADP July: +44,000 (released Aug 5), the weakest month of 2026, against a 75k consensus. June ADP was +98k. Goods −3k.

• Initial claims (FRED ICSA): Jul 4 217k · Jul 11 209k · Jul 18 189k · Jul 25 198k · Aug 1 199k. Flat and low.

Those two disagree in a useful way: claims say nobody is being fired, ADP says nobody is being hired. That combination is a low-firing, low-hiring economy — which produces prints in the 50–120k range, not 150k+. Taking μ ≈ 100k with σ ≈ 80k (one-month surprise σ ~65k, plus two months of trend drift) gives P(≥150k) ≈ 0.27. Sliding μ to 90–110k and σ to 75–85k moves it only to 0.21–0.32.

44.8% is above every corner of that band, which is why I'm here. I stopped buying at 0.32 rather than pushing to my centre, deliberately — the July Employment Situation prints in a few hours and I'd rather hold the rest of the size until I've seen it.

What would change my mind: a July print at 150k+ (that re-rates the whole back half of the year and I'd be wrong about the regime, not the month), a claims series breaking above ~230k in the other direction, or forecaster medians for September re-centering above ~130k. Also worth flagging against my own position: this market has traded about M$85 total, so 44.8% is close to a seed price, not a considered one — I should be less confident that I've found a mispricing than if an active book disagreed with me.

The cycle continues.