--
Loading current read...
Primary date
--
Equal score
--
Risk rank
--
Flag breadth
--
Current Decile Output
Historical ex-ante outcomes for months matching the current 75/25 score-plus-breadth decile.
| Asset | Matched Months | Avg 12M Return vs Base | 12M Vol vs Base | Avg 12M Max DD vs Base | 12M DD <= -10% vs Base | 12M Return <= -10% vs Base | 12M Return < 0 vs Base |
|---|
Current Methodology
A 75/25 rank model: 75% four-factor score percentile plus 25% breadth percentile, read through equal-sized deciles.
Historical Drawdown Map
Broad-market S&P 500 drawdowns, NBER recessions, and the frozen 75/25 score-plus-breadth risk rank.
| Event | Peak | Crossed -10% | Trough | Depth | Pre-Signal | Max Prior Rank | Ex-Post Catalyst |
|---|
Risk Rank And Thresholds
Primary 75/25 percentile rank compared with top-quintile and top-decile cutoffs.
Four Inputs
ERP tightness, Baa tightness, broad non-broker stock saturation, and broker-customer leverage on comparable rolling-z scales.
Tail-Band Cross-Check
Secondary diagnostic bands from score thresholds and breadth. The primary model is the decile rank above.
| Regime | Asset | Months | Share | Avg 12M Return | Avg 12M DD | 12M DD <= -10% | 12M Return <= -10% | 12M Return < 0 | Dual Score |
|---|
Individual Factor Checks
Each factor tested alone using the same live threshold discipline.
| Factor | Flagged | 12M DD Hit | DD AUC | 12M Loss Hit | Loss AUC |
|---|
Primary Decile Calibration
Equal-sized historical buckets for the 75/25 score-plus-breadth rank. This is the main empirical read.
| Quintile | Months | Avg Rank | Avg Score | Avg Breadth | Avg 12M Return | 12M DD <= -10% | 12M Return <= -10% | Dual Score |
|---|
| Decile | Months | Avg Rank | Avg Score | Avg Breadth | Avg 12M Return | 12M DD <= -10% | 12M Return <= -10% | Dual Score |
|---|
Breadth Backup
The same model collapsed to active factor count. This is a confirmation layer, not the primary score.
| Flags | Months | Avg 12M Return | 12M DD <= -10% | 12M Return <= -10% |
|---|
Pure Score Deciles
Diagnostic: same four factors without the 25% breadth rank. Kept for comparison because the variants were close.
| Decile | Months | Avg Score | Avg Breadth | Avg 12M Return | 12M DD <= -10% | 12M Return <= -10% | Dual Score |
|---|
Factor Overlap / Independence
Tests whether the four flags behave like distinct mechanisms or duplicated risk-appetite signals.
| Pair | N | Pearson | Spearman |
|---|
| Flag Pair | Both | P(Right | Left) | Phi |
|---|
| Exact Flags | Months | Avg 12M Return | 12M DD <= -10% | 12M Loss <= -10% |
|---|
| Model | Target | N | Top Quintile | AUC |
|---|
Data Freshness
The primary read is capped where all four primary factors are actually observed.