Methodology
This page shows, without hiding anything, how we validate our rules and what their limits are.
'Market now' describes the current state, not a forecast
The old verdict tried to say whether the market would be up or down a month later. Tested fairly on past data the model had not seen, its balanced accuracy was 48-49% in all four markets (Bitcoin 47.9%) - about a coin flip.
We also built a model to forecast next month's swings. It did no better than simply saying 'the current state continues' (difference -4.4 to +2.0 pts, not statistically distinguishable).
So from 2026-09-22 the screen is not a forecast. It describes how much the market is swinging now compared with usual. No hit rate is attached.
- Quiet - recent swings below 0.8x usual (clears at 0.9x or above).
- Normal - moving about as usual.
- Volatile - recent swings at 1.3x usual or more (clears below 1.15x).
- Falling sharply - more than 12% below the 60-day high with swings at 1.4x usual or more, or a one-day drop of more than 4% (and more than 4x a usual day) in the last 5 sessions. KOSDAQ uses 18%; Bitcoin uses 20% and 8% a day.
Swings = standard deviation of daily log returns over the last 20 trading days. Usual = the median of that value over the past year (365 days for Bitcoin). To avoid flip-flopping, the bar to leave a state is looser than the bar to enter it.
Applied to history (US since 1991, KOSPI 1998, KOSDAQ 2001, Bitcoin 2015), the state changes about once a month. The month after a quiet day swung about 0.8x usual; after a volatile day about 1.2x. It is a factual description that recent swings tend to last a few weeks.
It does not say whether prices will rise or fall. Not investment advice.
How the previous method (direction forecast) tested
| Market | Balanced accuracy, up & down days (%) | Always saying 'up' (%) | Swing forecast − simple persistence (pp, 95% range) |
|---|---|---|---|
| US | 49.3 | 64.9 | +0.4 (-2.6 ~ +3.1) |
| KOSPI | 49.1 | 57.8 | +1.1 (-2.2 ~ +4.2) |
| KOSDAQ | 48.8 | 54.8 | -4.4 (-8.3 ~ -0.5) |
| Bitcoin | 47.9 | 52.3 | +2.0 (-4.1 ~ +7.2) |
The current market state applied to history
| Market | Period | Quiet / Normal / Volatile / Falling sharply (% of days) | Swings a month later (x usual, same order) | Changes per month |
|---|---|---|---|---|
| US | 1991-01-28 ~ 2026-09-18 | 35.4 / 37.4 / 20.8 / 6.4 | 0.80 / 0.97 / 1.24 / 1.62 | 0.91 |
| KOSPI | 1998-01-30 ~ 2026-09-18 | 31.2 / 42.4 / 17.2 / 9.2 | 0.83 / 0.99 / 1.17 / 1.45 | 1.03 |
| KOSDAQ | 2001-11-22 ~ 2026-09-18 | 34.6 / 34.7 / 21.2 / 9.5 | 0.83 / 0.99 / 1.21 / 1.29 | 1.23 |
| Bitcoin | 2015-10-06 ~ 2026-09-19 | 35.7 / 35.5 / 17.8 / 11.0 | 0.83 / 0.94 / 1.20 / 1.27 | 1.07 |
Record archive
Previous-method (direction forecast) records are kept and still scored by the 21-trading-day rule. Calendar: US · Korea · Crypto · commodity
How we validate
- Walk-forward: scores use only data available up to the rebalance date, and results are checked only against what actually happened afterward.
- Out-of-sample testing: performance is measured on periods not used to set the rules.
- No lookahead: data not yet available at that point in time (e.g. the next day's close) is never used in the calculation.
- Transaction cost assumptions: minimum validation window 3 years, commission 0.005/share, minimum commission 1.0, slippage 5bp, minimum paper-trading period 90 days.
Survivorship bias
Backtesting only with stocks that are alive today inflates results, since delisted stocks are excluded. So the goal is to store the universe (candidate list) as a weekly snapshot of what actually existed at that time. The buy-fitness table doesn't have that snapshot yet (partial coverage), so it uses today's universe even for past periods — stocks that were delisted or newly added in between aren't reflected.
Rules & parameters (fixed in advance)
The numbers below were fixed when the strategy was designed, not fitted after looking at the data. By default, buy candidates exclude companies with a caution signal in revenue, profit, or debt (e.g. ongoing losses, high debt) — this can be turned off per strategy.
| Parameter | value |
|---|---|
| top_k | 20 |
| horizon_days | 21 |
| max_position_pct | 10.0% |
| weight · trend_120 | 1.0 |
| weight · momentum_12_1 | 1.0 |
| weight · zscore_20 | -0.5 |
| weight · vol_regime | -0.5 |
| weight · dist_high_252 | 0.5 |
| exit.balanced | stop 8% · trail 10% · take-profit 25% · max 120d |
| exit.trend | stop 10% · trail 15% · take-profit 40% · max 180d |
| exit.dip | stop 6% · trail 8% · take-profit 15% · max 60d |
| exit.stable | stop 5% · trail 7% · take-profit 12% · max 90d |
Parameter sensitivity
We check that small changes to the stop-loss level or the trend lookback window don't swing the results wildly (an overfitting check).
US
| Parameter | Avg return | Hit rate | Sample |
|---|---|---|---|
| trend_120 = 100d | +1.87% | 56% | 107 |
| trend_120 = 120d | +2.09% | 57% | 107 |
| trend_120 = 150d | +2.27% | 57% | 107 |
| stop = 6% | +0.73% | 46% | 2140 |
| stop = 8% | +1.04% | 51% | 2140 |
| stop = 10% | +1.18% | 53% | 2140 |
Korea
| Parameter | Avg return | Hit rate | Sample |
|---|---|---|---|
| trend_120 = 100d | +2.06% | 48% | 104 |
| trend_120 = 120d | +2.28% | 49% | 104 |
| trend_120 = 150d | +2.25% | 48% | 104 |
| stop = 6% | +1.09% | 33% | 2080 |
| stop = 8% | +1.42% | 38% | 2080 |
| stop = 10% | +1.81% | 43% | 2080 |
Crypto
| Parameter | Avg return | Hit rate | Sample |
|---|---|---|---|
| trend_120 = 100d | +2.70% | 44% | 101 |
| trend_120 = 120d | +2.81% | 44% | 101 |
| trend_120 = 150d | +2.69% | 44% | 101 |
| stop = 6% | +2.57% | 29% | 1967 |
| stop = 8% | +2.50% | 31% | 1967 |
| stop = 10% | +2.65% | 34% | 1967 |
Buy fitness
In past walk-forward validation, the share of stocks that rose 21 trading days (about a month) later when a (market verdict stage) × (that day's rank bucket) × (caution flag present/absent) condition held. No arbitrary weights. Verdict-stage labels are today's rules applied retroactively to the whole history (an approximation — those rules didn't necessarily exist back then).
US
| Verdict stage | Rank bucket | Caution flag | Share up | Avg return | Sample (stock-days) | Sample (trading days) |
|---|---|---|---|---|---|---|
| Risk | 1-5 | No | 57.5% | +3.00% | 749 | 167 |
| Risk | 1-5 | Yes | 53.5% | +5.73% | 86 | 57 |
| Risk | 6-10 | No | 54.7% | +2.32% | 697 | 167 |
| Risk | 6-10 | Yes | 53.6% | +0.69% | 138 | 91 |
| Risk | 11-20 | No | 57.8% | +2.43% | 1387 | 167 |
| Risk | 11-20 | Yes | 58.0% | +1.85% | 283 | 115 |
| Risk | 21-50 | No | 56.8% | +1.71% | 3988 | 167 |
| Risk | 21-50 | Yes | 56.9% | +2.78% | 1022 | 157 |
| Caution | 1-5 | No | 53.5% | +0.95% | 5300 | 1210 |
| Caution | 1-5 | Yes | 53.7% | +0.76% | 775 | 483 |
| Caution | 6-10 | No | 53.5% | +0.68% | 4956 | 1211 |
| Caution | 6-10 | Yes | 54.9% | +0.85% | 1119 | 656 |
| Caution | 11-20 | No | 53.3% | +0.65% | 9631 | 1215 |
| Caution | 11-20 | Yes | 56.6% | +0.94% | 2519 | 915 |
| Caution | 21-50 | No | 53.7% | +0.69% | 28358 | 1215 |
| Caution | 21-50 | Yes | 54.7% | +0.64% | 8092 | 1153 |
| Mixed | 1-5 | No | 63.7% | +3.38% | 3546 | 764 |
| Mixed | 1-5 | Yes | 57.7% | +2.31% | 284 | 184 |
| Mixed | 6-10 | No | 60.4% | +2.75% | 3425 | 763 |
| Mixed | 6-10 | Yes | 58.8% | +2.48% | 405 | 261 |
| Mixed | 11-20 | No | 60.1% | +2.54% | 6685 | 766 |
| Mixed | 11-20 | Yes | 58.3% | +2.39% | 975 | 402 |
| Mixed | 21-50 | No | 60.1% | +2.41% | 19934 | 766 |
| Mixed | 21-50 | Yes | 57.7% | +1.41% | 3046 | 582 |
| Good | 1-5 | No | 70.1% | +4.33% | 402 | 83 |
| Good | 1-5 | Yes | 68.9% | +4.04% | 415 | 83 |
| Good | 6-10 | No | 62.9% | +2.92% | 380 | 83 |
| Good | 6-10 | Yes | 60.0% | +2.90% | 35 | 28 |
| Good | 11-20 | No | 63.7% | +3.68% | 758 | 83 |
| Good | 11-20 | Yes | 66.7% | +4.80% | 72 | 40 |
| Good | 21-50 | No | 67.9% | +4.60% | 2278 | 83 |
| Good | 21-50 | Yes | 55.7% | +1.38% | 212 | 64 |
Korea
| Verdict stage | Rank bucket | Caution flag | Share up | Avg return | Sample (stock-days) | Sample (trading days) |
|---|---|---|---|---|---|---|
| Risk | 1-5 | No | 44.7% | +2.86% | 94 | 20 |
| Risk | 1-5 | Yes | 45.0% | +2.80% | 100 | 20 |
| Risk | 6-10 | No | 51.1% | +3.56% | 92 | 20 |
| Risk | 6-10 | Yes | 52.0% | +3.77% | 100 | 20 |
| Risk | 11-20 | No | 46.0% | +4.83% | 174 | 20 |
| Risk | 11-20 | Yes | 48.0% | +5.24% | 200 | 20 |
| Risk | 21-50 | No | 49.1% | +2.37% | 507 | 20 |
| Risk | 21-50 | Yes | 50.5% | +6.38% | 93 | 19 |
| Caution | 1-5 | No | 47.8% | +0.86% | 1501 | 328 |
| Caution | 1-5 | Yes | 55.4% | +7.47% | 139 | 111 |
| Caution | 6-10 | No | 45.6% | +0.60% | 1415 | 328 |
| Caution | 6-10 | Yes | 48.0% | +4.72% | 225 | 162 |
| Caution | 11-20 | No | 47.0% | +1.19% | 2812 | 328 |
| Caution | 11-20 | Yes | 47.4% | +2.28% | 468 | 250 |
| Caution | 21-50 | No | 45.3% | -0.12% | 8543 | 328 |
| Caution | 21-50 | Yes | 44.1% | +1.61% | 1285 | 309 |
| Mixed | 1-5 | No | 47.7% | +1.74% | 4842 | 1073 |
| Mixed | 1-5 | Yes | 54.0% | +7.21% | 517 | 361 |
| Mixed | 6-10 | No | 46.7% | +2.13% | 4640 | 1070 |
| Mixed | 6-10 | Yes | 48.2% | +3.28% | 710 | 496 |
| Mixed | 11-20 | No | 48.9% | +1.93% | 9005 | 1070 |
| Mixed | 11-20 | Yes | 46.9% | +2.22% | 1685 | 768 |
| Mixed | 21-50 | No | 47.5% | +0.77% | 26774 | 1068 |
| Mixed | 21-50 | Yes | 48.8% | +2.17% | 5220 | 1011 |
| Good | 1-5 | No | 51.3% | +3.54% | 3381 | 743 |
| Good | 1-5 | Yes | 58.7% | +4.27% | 334 | 254 |
| Good | 6-10 | No | 54.0% | +3.72% | 3214 | 739 |
| Good | 6-10 | Yes | 57.5% | +5.54% | 501 | 326 |
| Good | 11-20 | No | 54.6% | +3.87% | 6328 | 743 |
| Good | 11-20 | Yes | 52.0% | +4.02% | 1102 | 509 |
| Good | 21-50 | No | 55.5% | +3.64% | 18634 | 743 |
| Good | 21-50 | Yes | 53.1% | +4.70% | 3656 | 668 |
Crypto
| Verdict stage | Rank bucket | Caution flag | Share up | Avg return | Sample (stock-days) | Sample (trading days) |
|---|---|---|---|---|---|---|
| Risk | 1-5 | No | 36.2% | -2.62% | 188 | 41 |
| Risk | 1-5 | Yes | 37.1% | -1.94% | 205 | 41 |
| Risk | 6-10 | No | 29.0% | -4.72% | 183 | 41 |
| Risk | 6-10 | Yes | 29.3% | -4.44% | 205 | 41 |
| Risk | 11-20 | No | 31.7% | -6.45% | 357 | 41 |
| Risk | 11-20 | Yes | 19.6% | -7.21% | 51 | 14 |
| Risk | 21-50 | No | 13.8% | -11.90% | 145 | 36 |
| Risk | 21-50 | Yes | 13.1% | -12.62% | 160 | 39 |
| Caution | 1-5 | No | 43.3% | -0.01% | 4749 | 1256 |
| Caution | 1-5 | Yes | 50.5% | +11.45% | 640 | 366 |
| Caution | 6-10 | No | 43.7% | +0.31% | 3996 | 998 |
| Caution | 6-10 | Yes | 46.2% | +3.79% | 688 | 338 |
| Caution | 11-20 | No | 37.3% | -1.63% | 6424 | 797 |
| Caution | 11-20 | Yes | 38.6% | +2.86% | 1173 | 335 |
| Caution | 21-50 | No | 27.9% | -7.50% | 1429 | 448 |
| Caution | 21-50 | Yes | 19.2% | -10.08% | 255 | 116 |
| Mixed | 1-5 | No | 52.0% | +6.06% | 7936 | 1884 |
| Mixed | 1-5 | Yes | 53.7% | +9.11% | 1380 | 761 |
| Mixed | 6-10 | No | 50.1% | +4.95% | 7257 | 1768 |
| Mixed | 6-10 | Yes | 52.5% | +9.22% | 1270 | 656 |
| Mixed | 11-20 | No | 48.3% | +2.77% | 10416 | 1372 |
| Mixed | 11-20 | Yes | 51.7% | +10.01% | 1629 | 568 |
| Mixed | 21-50 | No | 39.5% | -1.89% | 2632 | 802 |
| Mixed | 21-50 | Yes | 46.2% | +4.18% | 379 | 200 |
| Good | 1-5 | No | – | – | 0 | 0 |
| Good | 1-5 | Yes | – | – | 0 | 0 |
| Good | 6-10 | No | – | – | 0 | 0 |
| Good | 6-10 | Yes | – | – | 0 | 0 |
| Good | 11-20 | No | – | – | 0 | 0 |
| Good | 11-20 | Yes | – | – | 0 | 0 |
| Good | 21-50 | No | – | – | 0 | 0 |
| Good | 21-50 | Yes | – | – | 0 | 0 |
Testing the company-financials filter
We re-ran the rule 'drop companies with risky financials from the candidate list' using only the figures that were already filed on each day (point-in-time — no later restatements, nothing not yet published). For every trading day we take the 20 candidates and compare their return 21 trading days later against the index. Because holding periods overlap, days are not independent, so the 95% interval comes from a 21-day block bootstrap (1,000 resamples). The bar for 'adopted' was set before looking at the results: the lower bound must be above 0 and the sign must agree in at least two half-year periods.
US
| Financial summary | 21-day mean | Share that rose | vs index | Cases (stock x day) |
|---|---|---|---|---|
| Solid | +5.16% | 55% | +4.00% | 732 |
| Average | +2.12% | 58% | +0.44% | 2938 |
| Caution | +2.02% | 51% | +0.15% | 1290 |
| No data | – | – | – | 0 |
| How candidates are picked | vs index | Difference vs default | 95% interval | Verdict |
|---|---|---|---|---|
| Default (by score) | +0.89% | – | – | – |
| Exclude caution companies | +1.27% | +0.38% | [-0.17%, +1.25%] | On hold (not enough evidence) |
| Solid companies first | +1.90% | +1.01% | [-0.24%, +2.51%] | On hold (not enough evidence) |
Korea
| Financial summary | 21-day mean | Share that rose | vs index | Cases (stock x day) |
|---|---|---|---|---|
| Solid | +7.61% | 63% | -0.57% | 826 |
| Average | +6.04% | 56% | -1.22% | 3314 |
| Caution | +2.73% | 55% | -4.16% | 854 |
| No data | -11.67% | 17% | -11.97% | 6 |
| How candidates are picked | vs index | Difference vs default | 95% interval | Verdict |
|---|---|---|---|---|
| Default (by score) | -1.63% | – | – | – |
| Exclude caution companies | -1.37% | +0.26% | [-0.33%, +0.96%] | On hold (not enough evidence) |
| Solid companies first | -1.71% | -0.08% | [-1.43%, +1.03%] | On hold (not enough evidence) |
Scoring the sharp-drop risk score
The sharp-drop alert is not a forecast - it is a daily thermometer for stress signs that often show up before big drops. Ten to twelve signals are turned into percentiles (0-1) using only data available up to that day, then combined by a model that is re-trained every January on prior data only, giving a 0-100 score. The stage bands (calm, watch, alert, danger) are chosen on the training window as well, and the table below re-scores only the period that was never used for training. We only have four years of data, so the scored window is short, and outside the US and Japan the danger stage did not separate real drops - treat it as a thermometer, not a signal.
US
| Stage (score) | Cases (days) | -7% rate | -10% rate | Avg max drawdown | vs base |
|---|---|---|---|---|---|
| Calm | 1108 | 5.6% | 1.8% | -2.25% | 0.63x |
| Watch | 412 | 7.5% | 2.4% | -2.22% | 0.85x |
| Alert | 349 | 15.2% | 7.7% | -2.91% | 1.71x |
| Danger | 92 | 30.4% | 16.3% | -6.91% | 3.43x |
Korea
| Stage (score) | Cases (days) | -7% rate | -10% rate | Avg max drawdown | vs base |
|---|---|---|---|---|---|
| Calm | 3296 | 11.1% | 5.1% | -3.03% | 0.77x |
| Watch | 1204 | 16.0% | 8.6% | -3.71% | 1.11x |
| Alert | 979 | 20.9% | 11.8% | -4.02% | 1.45x |
| Danger | 388 | 21.1% | 13.4% | -4.65% | 1.47x |
Crypto
| Stage (score) | Cases (days) | -7% rate | -10% rate | Avg max drawdown | vs base |
|---|---|---|---|---|---|
| Calm | 1869 | 36.3% | 25.5% | -6.69% | 0.94x |
| Watch | 499 | 45.1% | 32.1% | -8.02% | 1.17x |
| Alert | 426 | 41.1% | 26.1% | -6.88% | 1.06x |
| Danger | 64 | 37.5% | 25.0% | -5.93% | 0.97x |
How to read the statistics
- If the sample (n) is under 30, chance is too likely a factor — we don't draw conclusions.
- Don't look at average return alone; check hit rate and sample size together.
- Past performance does not guarantee future returns.
Limits & regimes where this doesn't work well
- Being rule-based, it reacts slowly to sudden news or policy shocks.
- For thinly traded stocks/markets, the slippage assumption may not match reality.
- Markets with a short history (e.g. crypto) have lower validation confidence.
- Many days in the forecast record were calculated retroactively, before launch (2026-09-06). Those weren't forecasts we actually published that day, so we exclude them from the scorecard's headline numbers and show them separately, labeled 'backfilled'.
- The KOSDAQ verdict inside the Korea market (added 2026-09-19) runs the same formula as KOSPI on the KOSDAQ index (^KQ11, Yahoo Finance). The sharp-drop alert is still KOSPI-index-only, and the KOSDAQ forecast record starts 2026-09-19 with no retroactive backfill.
Data sources & license
Price data comes from public market data providers and is provided only as statistical information, not investment advice.
Who makes this
MarketBaro is a side project built by a developer who also runs an AI startup in Singapore. It's not a company product, so we try to keep the rules and the data as transparent as possible.