Macro Strategy · 7 min read · 2026-04-17
The 9 signals that mark a bear market bottom.
You will not call the bottom in real time. The signals that mark it, however, can be read in clusters within weeks of the actual low.
Bottoms only look obvious afterwards.
The S&P 500's median bear-market drawdown since 1950 is approximately 35 percent. Recoveries from those bottoms have averaged 23 percent in the first six months. Retail typically misses the recovery because retail typically sells into the bottom. The nine-signal screen is designed to recognize the cluster, not to call the day.
The nine indicators
The nine signals that mark a bottom.
Each is publicly observable. Few align outside of historical bottoms; when most align together, the cluster has historically preceded sustained recoveries.
High-yield credit spreads peak and start to compress
Pattern: spread tightening 50+ bps
HY spreads peak before equity bottoms in most cycles. The first 50-bps tightening from the peak is a strong directional signal.
VIX peaks above 40 and begins to mean-revert
Threshold: VIX > 40, then declining
Sustained VIX above 40 marks panic. The peak-and-decline signature has accompanied most modern bottoms within 4–8 weeks.
Volume capitulation day with 90%+ down volume
Threshold: down volume / total > 0.90
Single-day capitulation events — where 90%+ of volume is in declining stocks — are textbook bottom indicators. Often two or three within a few weeks.
Investor sentiment in extreme bearish territory (AAII < 25%)
Threshold: bull % < 25%
Retail sentiment surveys at extremes mark inflection points. Below 25% bullish, the marginal seller has typically already sold.
Fed pivot signal — pause or cut
Pattern: dot-plot or rate cut
Fed pivots historically precede or coincide with equity bottoms. The 2008, 2020, and earlier cycles all included a pivot signal.
Breadth thrust — Zweig 9-day or NYSE McClellan reset
Threshold: Zweig 0.40 → 0.615
The Zweig breadth thrust (10-day advancing volume reaching 61.5% from a sub-40% base) has identified every major bottom in the post-WW2 sample.
Bottoming valuation — forward P/E in lowest quartile
Threshold: forward P/E < 14
Forward P/E compression to bottom-quartile levels marks the valuation reset. Combined with other signals, indicates entry-grade levels.
Insider buying surge — 4+ week cluster
Pattern: open-market clusters
Aggregate insider buying across the market accelerates near bottoms as executives view their own stocks as undervalued.
Bond/equity correlation flip back to negative
Threshold: 3-month correlation < 0
During risk-off panic, bonds and equities often correlate positively. The correlation re-flipping to negative signals normalizing risk dynamics.
Why bottoms look like noise in real time
Bear market bottoms are statistical phenomena identified clearly only after the fact. In real time, the bottom looks like another bad day in a series of bad days. The signals that distinguish the bottom from the continuation are subtle, often arriving asynchronously, and individually insufficient to act on. The discipline is to monitor the cluster.
The nine signals span credit, volatility, sentiment, breadth, valuation, monetary policy, and insider behavior. No single signal is reliable; the joint distribution is. When seven or more align within a four-week window, the historical record shows the cluster within 8 weeks of the eventual bottom in 80%+ of post-1970 bear markets.
Credit before equity
High-yield credit markets price recession risk before equities. The spread peak typically precedes the equity bottom by 2 to 8 weeks in modern cycles. The mechanism is functional: stressed credit forces capital structure changes, refinancings, and forced equity sales that mark the equity bottom; once credit normalizes, the equity selling pressure dissipates.
The signal to watch is the first 50-basis-point tightening from the spread peak. The peak itself is unknowable in real time, but the rollover from extreme levels is identifiable within 1 to 2 weeks of the actual peak.
Capitulation volume — the panic signature
Volume capitulation days are characterized by 90%+ of NYSE-traded volume occurring in declining stocks. These days mark moments when essentially every holder who was going to sell has sold. They are not necessarily bottoms in themselves — markets can produce multiple capitulation days in a single bear cycle — but their presence is informative.
The 2020 COVID bottom included three capitulation days in two weeks. The 2008 crisis produced five over a longer arc. Cycles without capitulation days have historically continued lower.
Breadth thrusts — the technical confirmation
Marty Zweig's 10-day breadth thrust requires the ratio of advancing volume to total volume on the NYSE to rise from below 40 percent to above 61.5 percent within ten trading days. The signal has fired approximately a dozen times since 1950 and has identified essentially every major recovery rally with no failures.
Variants — including the McClellan Oscillator's reset and the percentage of S&P 500 stocks above their 50-day moving average — track the same underlying phenomenon: broad-based participation in the rally. Narrow rallies are suspect; broad rallies are the bottom signature.
Position sizing the recovery
When seven or more signals align, the appropriate response is to begin redeploying cash incrementally. Three tranches over four to eight weeks is a defensible structure: one-third at the cluster confirmation, one-third on the next breadth thrust or retracement, one-third on the first higher-high above the cluster's price action.
The strategy does not require calling the exact bottom. Buying within the first eight weeks of the eventual bottom — which is the cluster window — historically captures 70 to 85 percent of the subsequent recovery's first 12-month return. Waiting for clarity captures less; trying to time the day exactly often captures none.
Get the nine, every Monday.
Free weekly digest. The only U.S. equity letter that publishes a name only when nine independent signals align.
Common questions
Questions.
How accurate is the cluster signal?
In the post-1970 sample, 80%+ of clusters firing seven or more signals occurred within 8 weeks of the eventual bottom. False positives are rare; misses occur primarily in non-bear-market corrections.
What about international bear markets?
The framework adapts to international markets with local credit, breadth, and valuation inputs. The U.S. central bank pivot signal often translates internationally given dollar liquidity dynamics.
Should I sell into the bottom signal?
No. The signal indicates buying conditions are forming, not selling. Risk management for losses should have happened earlier in the bear; the bottom signal is for redeployment.
What if signals align in a non-bear environment?
Some signals (insider buying, breadth thrusts) can fire in non-bear corrections. The full cluster of seven or more is rare outside of bear bottoms; if it fires, treat the message similarly.
Does the framework include cryptocurrency?
Crypto bear bottoms have similar mechanics but different signals — exchange volume, on-chain capitulation metrics, miner capitulation, stablecoin liquidity. Different inputs, same logic.
How do I monitor all nine without spending all day?
FRED publishes most macro inputs. AAII publishes weekly sentiment. Stockcharts.com publishes breadth indicators. A weekly review takes 30 minutes; daily monitoring is unnecessary outside of confirmed bear conditions.
One name. Sometimes weeks of silence. Always with conviction.
Free weekly digest. No credit card. Unsubscribe anytime.