Daily Macro US
Cleveland Nowcast
3.92%
FIFTH straight session — June CPI is baked; only rounding risk remains. Print Tuesday 8:30 ET
June CPI Market (pos-014)
20c
15 → 20c since entry; crowd now splits 3.8 (49c) / 3.7 (30c) — leaning soft from below
10Y Treasury
4.55%
TOUCHED ~4.60 Wednesday (7-wk high) — pos-017 needs just ~25bp now; FV 32 → 38
Fed Hike 2026 (pos-010)
52.5%
Pulled back from 58 WHILE futures price 34.4bp of hikes — FedWatch gap re-widens to ~10pp; deploy gate armed
Jobless Claims
215K
Lowest since May 23 — labor stabilizing, no-cut case intact (pos-013)
Equities
S&P 7,543
+0.81% Thu, Nasdaq +1.3% on chips; gold rebounds to ~$4,120
Four days to CPI, and the market is walking toward our bracket one grudging cent at a time. pos-014 (June CPI = 3.9%) trades at 20c — up from a 15c entry — while the Cleveland nowcast printed 3.92% for the FIFTH consecutive session. June's number is mathematically baked at this point: the model's remaining uncertainty is rounding, not direction. The crowd, having round-tripped through 4.0-4.2 and back, now splits its chips between 3.8 (48.9c) and — new this morning — 3.7 (29.7c, up from 5c). Someone is betting the oil crash pulled harder than the model thinks. We hold the mode at $75 and let Tuesday referee. The stealth story of the week is the 10-year: it TOUCHED ~4.60 on Wednesday — a seven-week high — and closed Thursday at 4.551. pos-017, entered yesterday at 16.5c on a 32bp-to-target thesis, now needs only ~25bp with five and a half months of runway, and the market has repriced it all of half a cent. FV raised 32 → 38; the correlation cap keeps us at $25, but this is the quiet compounder to watch. Meanwhile a familiar divergence re-opened: fed funds futures now price 34.4bp of hikes by year-end (NY Fed's Williams explicitly flagged AI-driven demand as an inflation risk, and jobless claims fell to 215K, the lowest since May), yet Polymarket's hike-2026 market PULLED BACK from 58 to 52.5c. That's the FedWatch gap re-widening to ~10pp — the same setup that paid us at 55c entry and again on the July convergence trade. The reserve deploy gate is now armed: a hot CPI print (≥0.3% MoM) with the market at or below 52c triggers the $100 add. Equities rallied Thursday (Nasdaq +1.3%, S&P +0.81% to 7,543) on chips; gold rebounded to ~$4,120 as yields eased off the highs; and the Hormuz postscript keeps writing itself — strait attacks flared again Thursday per Schwab's morning note, with the market we exited at 14.5c still languishing single-digit. Book: +$27.95 unrealized, +$576.28 realized, 9/10. Tuesday decides the flagship.
Today's Market Moves
June CPI Annual (pos-014)
17%→20%+3pp
Converging to the model: 15 → 17 → 20c this week while Cleveland sat immobile at 3.92. The new wrinkle is 3.7 bid at 29.7c — the crowd's second theory in three days, now missing LOW instead of high. The model has watched both theories come and go without moving a basis point. +$25 MTM; hold to Tuesday, no add at 20c (entry edge was the trade).
Fed Rate Hike 2026 (pos-010)
58%→52.5%-6pp
The gap re-widens: market 52.5c vs futures pricing 34.4bp of year-end hikes (implying ~63+). Williams talking AI-demand inflation and claims at 215K don't support a 5.5pp fade. DEPLOY GATE ARMED: CPI ≥0.3% MoM with price ≤52c → $100 add Tuesday. Floor 48c unchanged.
10Y Touches 4.8% (pos-017)
16.5%→17%+1pp
The 10Y touched 4.60 within 24 hours of our entry — a fifth of the required move, and the market repriced half a cent. Touch markets lag spot narratives; that lag IS the edge. FV 38.
China GDP Q2 (pos-016)
73%→67.5%-6pp
Drifted 5.5pp on nothing we can find — no NBS leak, no consensus shift in the wires. Uncomfortable but unactionable: the print lands Wednesday. No add into unexplained drift; conf stays 5.
Screening Table
| # | Market | Expiry | Market Price | Fair Value | Gap (pp) | Direction | Volume | Confidence |
|---|---|---|---|---|---|---|---|---|
| 1 | June CPI YoY = 3.9% | Jul 14 | 20% | 45% | +25pp | HOLD $75 YES — model immobile at 3.92, crowd on its second wrong theory | $$656K | 7/10 |
| 2 | 10Y Touches 4.8% Before 2027 | Dec 31 | 17% | 38% | +21pp | HOLD $25 YES — 10Y touched 4.60 Wed; correlation-capped | $$245K | 6/10 |
| 3 | Fed Rate Hike 2026 | Dec 2026 | 52.5% | 63% | +10pp | HOLD YES — gap re-widened; deploy gate armed (CPI ≥0.3% + ≤52c) | $$35K/day | 7/10 |
| 4 | BoJ 25bp Hike at Sep Meeting | Sep 2026 | 10% | 25% | +15pp | GATED — Jul 15 statement, 5 days | $Minimal | 5/10 |
| 5 | China GDP Q2 = 4.6-4.9% | ~Jul 15 | 67.5% | 85% | +17pp | HOLD $25 YES — unexplained 5.5pp drift, no add | $$144K | 5/10 |
| 6 | Zero Fed Cuts 2026 | Dec 31 | 77.6% | 85% | +7pp | HOLD YES — claims 215K support | $$24K/day | 7/10 |
Market vs Fundamentals
Market Price (red) vs Estimated Fair Value (green) — %
Top 5 Opportunities
1
June CPI YoY = 3.9% — YES
↑ BUY YES+25pp
Market price
20%
Fair value
45%
Gap: +25pp
The week in one position: entered at 15c against a crowd at 3.8, watched the crowd stampede to 4.0-4.2 and back, and now watch it try 3.7 — while the model prints 3.92 five days running. At 20c the market pays 4:1 that Cleveland's June arithmetic, with every major input already ingested, rounds where it says it rounds. Tuesday 8:30 ET ends the argument either way, and we'll publish the post-mortem whichever way it goes.
▵ Bull case
- Five consecutive 3.92 prints — the June nowcast no longer has inputs left to surprise it
- Crowd theories keep dying against the model (4.0-4.2 Tuesday, now 3.7)
- 20c still pays 4:1 on the base case
▿ Bear case
- 3.85-3.89 rounds to 3.9 but 3.84 doesn't — one hundredth matters
- The 3.7 bid at 29.7c is new money with a thesis (harder oil pass-through)
- Model error at 4-day horizon is small, not zero (±0.08 historically)
2
10Y Treasury Touches 4.8% Before 2027 — YES
↑ BUY YES+21pp
Market price
17%
Fair value
38%
Gap: +21pp
Entered Wednesday needing 32bp; the 10Y delivered 12 of them within 24 hours (4.60 touch, 7-week high) before easing. The market moved our contract from 16.5 to 17 — a half-cent reprice for a fifth of the distance. Touch markets systematically lag the paths that resolve them; we raised FV to 38 and would add if the book weren't already long this world four ways.
▵ Bull case
- ~25bp remaining with 5.5 months, a hiking Fed, heavy issuance, and Williams talking AI-demand inflation
- One hot CPI morning can print the touch by Wednesday
- Deepest apathy on the board — vol24 was $159 yesterday
▿ Bear case
- Soft CPI caps the summer path
- Correlation: the book's Fed-side positions win and lose together
- Yields eased Thursday — momentum isn't linear
3
Fed Rate Hike 2026 — YES
↑ BUY YES+10pp
Market price
52.5%
Fair value
63%
Gap: +10pp
The gap that started this position is back: futures price 34.4bp of year-end hikes while Polymarket fades to 52.5c. We've now been paid twice for trading this exact divergence toward the futures market. The deploy gate is armed and dated: hot CPI Tuesday with the market at or below 52c triggers the $100 reserve. Without the CPI confirmation, no add — the discipline is the strategy.
▵ Bull case
- Futures-implied probability ~63% vs 52.5c market
- Williams (AI demand) + claims 215K + split-but-hawkish minutes
- Sep and Oct windows both live
▿ Bear case
- The fade might be smart money front-running a soft CPI
- Third re-widening — at some point the market is telling us something
- Labor trend is still the bear's best card
4
BoJ 25bp Hike at September Meeting — YES
↑ BUY YES+15pp
Market price
10%
Fair value
25%
Gap: +15pp
Five days to the gate and the thin book keeps crawling our way — 6, 9, 10. Nothing to do until Wednesday's statement, which is precisely the point: the position we haven't taken has outperformed most positions we have.
▵ Bull case
- Yen still at extremes into the meeting
- Drift shows the book slowly agreeing with us
▿ Bear case
- Still can't absorb $25 cleanly
- Statement risk is binary
5
China GDP Q2 = 4.6-4.9% — YES
↑ BUY YES+17pp
Market price
67.5%
Fair value
85%
Gap: +17pp
The bracket drifted from 72.5 to 67.5 on nothing — no NBS leak we can find, no consensus revision in the wires. On paper the edge widened to 17pp; in practice, unexplained drift against a position is information of unknown sign. The rule for blind trades holds: original size, no averaging down, let Wednesday's print grade the thesis.
▵ Bull case
- Print regularity thesis unchanged
- Wider paper edge if the drift is noise
▿ Bear case
- Somebody sold 5.5pp of it — maybe they know the stimulus-print tail is live
- We remain blind on consensus