UNSUPERVISED · KMEANS · 5,745 DAYS · 2004–NOW
Market States
The machine grouped 22 years of trading days into five regimes — no labels given, just the shape of VIX, the yield curve, and S&P momentum & turbulence. Below: which regime today sits in, and the days that looked most like it.
TODAY · 2026-11-12
Range-bound calm
VIX
17.71
10y-2y
0.25pp
S&P 20d
-1.5%
S&P vol
22.18%
WHEN THE MARKET LAST LOOKED LIKE THIS — WHAT FOLLOWED
AVG +30 DAYS
-5.2%
AVG +90 DAYS
+6.8%
2018-11-16Range-bound calmVIX 18.1-10.5%+4.8%
2019-08-19Range-bound calmVIX 16.9+0.6%+10.8%
2018-11-15Correction / stressVIX 20-8.1%+3.8%
2018-11-19Correction / stressVIX 20.1-5.9%+6.6%
2019-08-29Range-bound calmVIX 17.9+1.6%+11.2%
2018-11-12Correction / stressVIX 20.5-8.7%+3.4%
Nearest neighbours in standardized feature space (the last 60 days excluded so matches aren't trivially recent). +30 / +90 columns are what the S&P 500 actually did after each date — history, not a forecast. Small sample; the future is not obliged to rhyme.
DOES THE SIGNAL HAVE AN EDGE? · S&P 500 forward return, 22 years
BASELINE · ANY DAY+3.5%72.9% hit · n=5655over the next 90 trading days, the bars below show distance from this
BY MARKET REGIME
Volatile recovery
+11.45%+7.95pp89.6% hit+8.85%…+16.35%n=270
Crash / capitulation
+4.61%+1.11pp51% hit-16.41%…+26.22%n=102
Range-bound calm● NOW
+3.48%-0.02pp≈ baseline · no edge+0.18%…+7.89%n=2526
Correction / stress
+3.2%-0.3pp≈ baseline · no edge-2.98%…+12.03%n=836
Steady climb (steep curve)
+2.48%-1.02pp70.4% hit-1.11%…+7.19%n=1921
BY VIX (FEAR)
< 15 · complacent
+2.75%-0.75pp75% hit0%…+6.45%n=2092
15–20 · normal● NOW
+2.97%-0.53pp≈ baseline · no edge-0.81%…+8%n=1786
20–30 · stressed
+3.59%+0.09pp≈ baseline · no edge-2.3%…+11.37%n=1328
≥ 30 · panic
+8.81%+5.31pp76.8% hit+1.47%…+18.54%n=449
BY YIELD CURVE (10Y–2Y)
inverted · < 0
+6.5%+3pp87.5% hit+2.93%…+9.73%n=786
flat · 0–0.5● NOW
+1.94%-1.56pp62.1% hit-2.77%…+6.75%n=1210
normal · 0.5–1.5
+3.02%-0.48pp≈ baseline · no edge+0.2%…+8.05%n=1879
steep · ≥ 1.5
+3.74%+0.24pp≈ baseline · no edge-1.17%…+9.3%n=1780
Forward returns are what actually happened, not a prediction. · Where a bucket's returns ≈ the baseline, the signal carries no usable edge. · Small samples (n < 30) are noisy — read the dispersion (p25–p75), not just the mean. · Past regimes are not obliged to rhyme with the future.
DOES THE TRADE TAB'S SIGNAL HAVE AN EDGE? · SET INDEX · 26 YEARS
What the conventional retail playbook says (RSI<30 = buy, lower BB = buy, pullback = buy) vs what 26 years of SET history actually shows. The bars are centred on the baseline — a bar at centre means no usable edge on this history.
BASELINE · ANY DAY (SET)+1.86%55.9% hit · n=612over the next 90 trading days
RSI(14) — is oversold/overbought predictive on SET?
The finding: SET rewards momentum continuation, not mean reversion.
< 30 · oversold● NOW
+1.31%-0.55ppmild drag-7.34…+12.07%n=92
30–45 · low
+0.62%-1.24ppmild drag-7.92…+10.68%n=192
45–60 · neutral
+2.86%+1ppmild edge-4.94…+11.01%n=172
60–70 · high
+1.89%+0.03pp≈ baseline-3.68…+8.52%n=72
> 70 · overbought
+3.21%+1.35ppmild edge-2.19…+9.2%n=84
EMA(9/21) CROSS — trend confirmation
EMA9 > EMA21 · bull
+2.01%+0.15pp≈ baseline-4.36…+9.24%n=248
EMA9 < EMA21 · bear● NOW
+1.76%-0.1pp≈ baseline-5.6…+11.46%n=364
BOLLINGER BAND POSITION (%B) — does 'touch the band' work?
Below lower band ≠ buy signal on SET history.
below lower band · %B < 0
+0.58%-1.28ppthin sample-9.12…+11.2%n=49
lower half · 0–0.5● NOW
+1.68%-0.18pp≈ baseline-5.51…+11.45%n=299
upper half · 0.5–1
+2.27%+0.41pp≈ baseline-4.4…+9.24%n=238
above upper band · %B > 1
+2.62%+0.76ppthin sample-3.22…+8.55%n=26
SET 20-DAY MOMENTUM — buy dips or ride momentum?
Sharp pullbacks show no edge. Strong rallies have the largest edge in this dataset.
< -8% · sharp pullback
+4.65%+2.79ppthin sample-3.92…+12.97%n=18
-8 to 0% · negative● NOW
+1.13%-0.73ppmild drag-6.42…+11.44%n=336
0–8% · positive
+2.48%+0.62ppmild edge-3.66…+9.72%n=211
> 8% · strong rally
+3.22%+1.36ppthin sample-0.76…+7.97%n=47
Forward returns are SET actual history, not a prediction. · RSI and BB signals are computed on the SET index, not individual stocks. · Where a bucket ≈ baseline it carries no usable edge on this history. · Small samples (n<30) are noisy — read dispersion (p25–p75), not just the mean. · The past need not rhyme with the future. · Yahoo did not serve the 26-year ^SET.BK history on this run. The sample is the longer durable series that cleared the minimum, not a retimestamp of the previous study.
WHAT STATISTICALLY PRECEDES THE SET? · GRANGER CAUSALITY · 112 DAYS
Granger causality is not a trading signal. It means that past values of a series helped predict SET returns in a linear regression over the historical sample — nothing more. Relationships break, markets adapt, and statistical significance in a backtest does not guarantee a live edge. Treat these as structural context for understanding what moves the SET, not as reasons to buy or sell.
STATISTICALLY SIGNIFICANT (p<0.05)
p<0.01Bitcoin (risk appetite)
F=16.57lag 1dp=0.0001
p<0.01Yield curve spread (10y−2y)
F=10.12lag 1dp=0.0016
NOT SIGNIFICANT — shown for completeness
p<0.10WTI Crude (energy)
F=3.31lag 1dp=0.0695
p<0.10USD/JPY (yen carry trade)
F=2.19lag 3dp=0.0889
p<0.10Copper (global growth proxy)
F=1.65lag 10dp=0.0922
not sig.S&P 500 (US equities)
F=2.00lag 3dp=0.1132
not sig.Nikkei 225 (Japan)
F=1.52lag 1dp=0.2184
not sig.CBOE VIX (fear gauge)
F=1.24lag 10dp=0.2623
not sig.Hang Seng (HK/China)
F=1.18lag 10dp=0.3018
not sig.US 10Y yield change
F=0.69lag 1dp=0.4075
not sig.USD Index (dollar strength)
F=0.64lag 1dp=0.4243
not sig.Gold (safe haven)
F=0.62lag 1dp=0.4331
not sig.USD/THB (baht weakness)
F=0.59lag 1dp=0.4446
not sig.US 2Y yield change
F=0.07lag 1dp=0.7943
Granger = predictive precedence, not true causation. ADF-tested for stationarity and aligned by trading session. · run 2026-11-09◐CACHED · —
ANOMALY WATCH · PRICE VS FUNDAMENTALS
IsolationForest over 25 SET names clustered on 14 fundamentals+price features. A flag means the stock trades in a way its cluster's numbers can't explain — the market is telling a story the statements don't carry yet. It is a pointer for looking, never a buy/sell call.
NAMECLUSTERANOMALY 0–120D RET
DELTA.BKgrowth_premium1.00-27.4%
ADVANC.BKgrowth_premium0.75+3.0%
BGRIM.BKspeculative0.67+18.6%
PTTGC.BKspeculative0.59-23.2%
mined 2026-11-12T18:37:39.875040 · refreshed weekly by ingestion/pattern_mine.py
THE FIVE REGIMES · share of 22 years
Range-bound calm← TODAY45.5%
VIX ~15.25 · 10y-2y 0.26pp · S&P 20d +1.93% · vol 11.45%
Steady climb (steep curve)33.4%
VIX ~16.71 · 10y-2y 1.91pp · S&P 20d +1.62% · vol 12.3%
Correction / stress14.6%
VIX ~26.62 · 10y-2y 0.87pp · S&P 20d -4.67% · vol 23.17%
Volatile recovery4.7%
VIX ~32.49 · 10y-2y 1.59pp · S&P 20d +5.78% · vol 32.51%
Crash / capitulation1.8%
VIX ~57.02 · 10y-2y 1.7pp · S&P 20d -13.74% · vol 68.37%
LAST 252 TRADING DAYS · regime tape
Steady climb (steep curve)Correction / stressRange-bound calmVolatile recoveryCrash / capitulation
KMeans (k=5) on z-scored FRED daily features; analogue = nearest neighbour in feature space, forward S&P 500 return reported · generated 2026-11-12. Credit spreads are excluded from the state model (FRED truncates the free series) but stay live in the Mr. Market gauge. See Data Integrity.