Regime analog — 2007 / 2022 / today

Static analysis page; the TODAY column is filled live from /status.json at view time — it refreshes with every daily deploy without code changes. Live values loading…

Cycle fingerprint

Feature20072022Today (live)
10Y at multi-decade high No — peaked ~5.3% (Jun 2007), mid-range for the post-1990 regime Building — 0.5% → 4.25% by Oct 2022, highest since 2007
Fed hiking into it Yes — 17 hikes 2004–06 to 5.25%, on hold into the Oct 2007 peak Yes — 0 → 4.50% in nine months, fastest cycle since Volcker Yes — Sep 2026 hike to 3.75–4.00%, median dot 4.125%+
Aggregate HY near record tights, with CCC cracked Yes — HY OAS ~260 bps (Jun 2007) near record tights; CCC already leaking wider No — HY widened with equities to ~600 bps; no bifurcation
ISM / claims strong at the peak Yes — ISM ~52 and claims ~300k at the peak; labor confirmed nothing Yes — claims ~200k at the Jan 2022 peak
Refinancing wall ARM resets into 5%+ mortgage rates (2006–08); lagging, then detonated Muted — termed-out corporate debt; stress surfaced in duration (gilts, SVB) HY wall 2026–28: refis into ~8–9% coupons against ~4% legacy
Oil-supply inflation overlay Demand-led — oil $60 → $140 into 2008; CPI ~4% Yes — Brent > $120 after the invasion; CPI 9.1% Oil > $100, backwardated
Fiscal-driven term premium No — deficit ~1% of GDP; term premium compressed Emerging — QT plus the gilt-crisis preview Yes — 30Y 5.34%; deficits ~6–7% of GDP funding the long end

Historical cells are rounded recollections for pattern-matching, not precise series values. TODAY cells marked live are read from the daily monitor output.

Verdict

2007 is the skeleton (cycle position: yield-wall, record-tight bifurcating credit, lagging refi wall, strong lagging labor data); 2022 is the skin (supply-side inflation, positive stock-bond correlation, Fed hiking into strength). Forward base case: 2007 sequencing with 2022 stickiness — yields exhaust first, credit velocity fires 1–3 months later, but the duration rally stalls nearer 4.25% than 3.80%. Tail case: 1973–74 stagflation (see /backtest-70s/) — the one episode class no equity-peak monitor catches; controlled via contango-flip exit, 5.40% invalidation, and the 40% carry/TIPS/gold block.

Why this supports the production monitor

The episode class most like today (credit-led systemic) is the class the v1 monitor historically catches with lead — that is the basis for keeping v1 frozen as the trading system while v2/v3 remain research tracks.