Credit Spreads

A credit spread is the extra yield lenders demand to hold a company’s bond instead of a Treasury — literally the market price of “will this company survive?”. When spreads are tight, credit sees blue skies; when they widen, the people paid to worry about default are worrying. Because credit investors face ruin rather than missed upside, they turn cautious before equity investors do — which makes this page an early-warning system you can use without ever touching a bond. Everything here reduces to one question: is credit confirming what equities are doing, or calling their bluff?

• As a risk governor (daily) — the traffic light sets your maximum size. Green zones: trade your plan. Amber: halve it. Red: defence. This single habit — letting credit cap your exposure — is the cheapest drawdown protection that exists.
• As a dip-buying filter — equity dip + spreads tightening = historically buyable. Equity dip + spreads still widening = falling knife. The divergence panel makes this check a two-second glance.
• As a rally lie-detector — 'Bearish divergence' — equities up, spreads widening — is the classic pre-correction signature. It doesn't time the top to the day, but rallies that credit refuses to fund have historically been the ones that fail.
• As a bottom-spotter — the feed currently serves only 3y (since 2023-10-31). No Stress/Crisis episodes are in that sample, so the base-rates table cannot yet show how panic bottoms behaved. Use the playbook's turn rule — buy the first sustained tightening from the peak, never the free-fall — not the table.

• Traffic light (HY OAS) — the headline high-yield spread with its percentile (within the window the feed serves) and zone. This is THE number people mean when they say 'credit spreads'.
• Velocity alarm — the speed of widening matters more than the level. +50bps in 20 sessions has historically accompanied the sharpest equity drawdowns — that threshold trips the alarm.
• CCC minus HY (early-warning wire) — stress always enters at the junkiest end of the market first. When CCC deteriorates faster than broad HY, the cycle is cracking from the bottom — often quarters before IG notices.
• Quality ladder — IG → BBB → HY → CCC side by side with percentile bars. Watch how far up the ladder the red has climbed.
• Credit vs equities overlay — the S&P against HY spreads (inverted, so confirming markets move together). The lines peeling apart IS the signal.
• Base rates + playbook — forward-return statistics per zone over the history the data feed serves — currently 3y (since 2023-10-31) — with thin rows flagged, and the concrete do/avoid list for wherever we are today.

Spreads are the real ICE BofA option-adjusted spreads from FRED (high yield BAMLH0A0HYM2, investment grade BAMLC0A0CM, BBB BAMLC0A4CBBB, CCC-and-below BAMLH0A3HYC) — the real thing, not ETF-price proxies. FRED now serves only a rolling ~3-year window of these ICE series (currently 3y (since 2023-10-31)), so percentiles are relative to that window, not to decades of history; the span is shown on each panel. Zone boundaries (300 / 400 / 550 / 750bps) are fixed from the longer pre-2023 ICE history and are not recomputed from the window. The forward-return base rates sample the same window every three sessions and measure S&P 500 returns over the following 63 trading days per zone; rows with fewer than 30 samples are flagged thin. The divergence state compares 21-day changes in the S&P and HY OAS: S&P up more than 0.5% with HY within +5bps is a confirming rally, +5 to +15bps an unconfirmed rally (amber), more than +15bps a bearish divergence; S&P down more than 0.5% with HY tighter than −5bps is a bullish divergence, within ±5bps “holding under selloff” (amber), wider an aligned risk-off; an S&P move inside ±0.5% is a flat tape. Everything recomputes on each 15-minute revalidate.

Credit spreads sit in Space and are the specific “tell” the Business Cycle read watches for: the first genuine, fast widening is usually the signal that a late-cycle regime is transitioning. Defensive posture should follow credit, not lag it — and offensive posture should wait for credit’s confirmation. Pair the traffic light with the Yield Curve’s regime quadrant: when both turn hostile at once, that is the environment where Motion and Time setups get overruled.