Claim check: Bounce after three down days
“After three down days in a row, the market bounces.”
What the data says
Does not hold Not supported by the data: after “Three down closes in a row”, the average 3-day return was 0.0% against +0.1% for all days (medians over 10 markets); it was higher than usual in 3 of 10 markets and statistically clear, with at least 20 independent events and the same sign in both halves, in 0. The sign mostly points the other way, and the result cannot be told apart from chance.
No effect we can tell from chance, or the opposite effect.
Who says it
Larry Connors's short-term mean reversion research.
Origins are as commonly cited and were not independently verified.
How we tested it
Event: three consecutive daily closes lower than the previous close. We record the return over the next three trading days (events at least 3 days apart) and compare it with the market's average 3-day return. The same pattern above the 200-day average is shown for contrast.
Results
Three down closes in a row 3-day (claim: higher than average)
| Market | Events | Mean after | All days | Difference | Median after | Up share | t-stat | Reading |
|---|---|---|---|---|---|---|---|---|
| S&P 500 | 223 | +0.3% | +0.2% | +0.2% | +0.4% | 58% | 1.3 (likely noise) | Right way, likely noise |
| Gold | 234 | -0.1% | +0.1% | -0.3% | -0.1% | 46% | -2.1 | Opposite, significant |
| Silver | 230 | -0.4% | +0.1% | -0.6% | -0.1% | 49% | -2.3 | Opposite, significant |
| Oil | 260 | -0.1% | 0.0% | -0.1% | 0.0% | 50% | -0.3 (likely noise) | No effect |
| Natural gas | 311 | 0.0% | -0.2% | +0.2% | -0.1% | 49% | 0.8 (likely noise) | Right way, likely noise |
| Commodities | 271 | 0.0% | 0.0% | 0.0% | +0.2% | 54% | -0.1 (likely noise) | No effect |
| Agriculture | 280 | -0.1% | 0.0% | -0.2% | -0.1% | 44% | -1.8 (likely noise) | No effect |
| EUR/USD | 307 | 0.0% | 0.0% | +0.1% | 0.0% | 51% | 1.1 (likely noise) | Right way, likely noise |
| Bitcoin | 266 | +0.5% | +0.5% | 0.0% | +0.6% | 58% | -0.1 (likely noise) | No effect |
| Ether | 197 | -0.1% | +0.5% | -0.6% | +0.3% | 54% | -1.0 (likely noise) | No effect |
Difference from the average, by market (3-day)
By horizon: mean difference from all days
| Market | 1-day | 3-day | 5-day |
|---|---|---|---|
| S&P 500 | +0.2% (353) | +0.2% (223) | +0.2% (207) |
| Gold | -0.1% (384) | -0.3% (234) | -0.2% (208) |
| Silver | -0.2% (381) | -0.6% (230) | -0.5% (203) |
| Oil | 0.0% (444) | -0.1% (260) | -0.1% (227) |
| Natural gas | 0.0% (541) | +0.2% (311) | +0.4% (272) |
| Commodities | 0.0% (454) | 0.0% (271) | -0.1% (239) |
| Agriculture | -0.1% (499) | -0.2% (280) | -0.1% (247) |
| EUR/USD | 0.0% (527) | +0.1% (307) | 0.0% (267) |
| Bitcoin | +0.2% (431) | 0.0% (266) | -0.3% (242) |
| Ether | -0.1% (318) | -0.6% (197) | -1.0% (174) |
Average -day return after the signal vs all days (S&P 500)
Three down closes, above the 200-day average · For comparison 3-day (claim: higher than average)
| Market | Events | Mean after | All days | Difference | Median after | Up share | t-stat | Reading |
|---|---|---|---|---|---|---|---|---|
| S&P 500 | 162 | +0.3% | +0.2% | +0.2% | +0.4% | 58% | 1.2 (likely noise) | Right way, likely noise |
| Gold | 100 | -0.1% | +0.1% | -0.2% | -0.1% | 46% | -0.9 (likely noise) | No effect |
| Silver | 72 | -0.5% | +0.1% | -0.6% | 0.0% | 50% | -1.4 (likely noise) | No effect |
| Oil | 84 | +0.3% | 0.0% | +0.2% | +0.2% | 55% | 0.6 (likely noise) | Right way, likely noise |
| Natural gas | 55 | -0.2% | -0.2% | 0.0% | -0.1% | 45% | 0.0 (likely noise) | Right way, likely noise |
| Commodities | 103 | +0.1% | 0.0% | 0.0% | +0.2% | 53% | 0.1 (likely noise) | Right way, likely noise |
| Agriculture | 100 | -0.2% | 0.0% | -0.2% | -0.2% | 43% | -1.2 (likely noise) | No effect |
| EUR/USD | 112 | 0.0% | 0.0% | 0.0% | -0.1% | 49% | 0.4 (likely noise) | Right way, likely noise |
| Bitcoin | 134 | +1.1% | +0.5% | +0.6% | +0.7% | 60% | 1.3 (likely noise) | Right way, likely noise |
| Ether | 72 | +1.9% | +0.5% | +1.4% | +1.3% | 65% | 1.6 (likely noise) | Right way, likely noise |
Stability across the two halves of history
A saying that only works in one half of the history is not much of a rule. Each row shows the effect in the full history, the first half and the second half; green means the claimed effect was present.
| Market | Full history | First half | Second half |
|---|---|---|---|
| S&P 500 | +0.2% (223) | +0.1% (104) | +0.3% (119) |
| Gold | -0.3% (234) | -0.3% (129) | -0.2% (105) |
| Silver | -0.6% (230) | -0.3% (117) | -0.8% (113) |
| Oil | -0.1% (260) | -0.3% (142) | +0.1% (118) |
| Natural gas | +0.2% (311) | +0.1% (161) | +0.4% (150) |
| Commodities | 0.0% (271) | -0.1% (150) | +0.1% (121) |
| Agriculture | -0.2% (280) | -0.2% (152) | -0.2% (128) |
| EUR/USD | +0.1% (307) | +0.1% (146) | 0.0% (161) |
| Bitcoin | 0.0% (266) | 0.0% (114) | 0.0% (152) |
| Ether | -0.6% (197) | -0.9% (91) | -0.2% (106) |
Context and cross-checks
Shown for context; not part of the verdict.
The site's own strategies
| Strategy | Market | Sharpe, 5 years | Benchmark, 5 years | Sharpe, full | Benchmark, full | CAGR, full | Benchmark, full | Max DD, full | Benchmark, full | Reading |
|---|---|---|---|---|---|---|---|---|---|---|
| MR-16 Connors RSI | SPY | 1.03 | 0.72 | 0.80 | 0.85 | +6.0% | +13.8% | -17% | -34% | No effect |
| MR-16 Connors RSI | GLD | 0.27 | 0.89 | 0.03 | 0.51 | 0.0% | +7.5% | -22% | -46% | No effect |
| MR-16 Connors RSI | USO | 0.73 | 0.68 | -0.13 | 0.05 | -3.7% | -4.8% | -73% | -95% | No effect |
| MR-16 Connors RSI | EUR/USD | -0.47 | -0.05 | -0.04 | -0.13 | -0.2% | -1.5% | -13% | -35% | Full history only |
| MR-16 Connors RSI | DBC | 0.24 | 0.63 | -0.25 | 0.21 | -2.1% | +2.2% | -41% | -66% | No effect |
| MR-16 Connors RSI | SLV | -0.23 | 0.57 | -0.16 | 0.35 | -2.6% | +6.1% | -56% | -76% | No effect |
| MR-16 Connors RSI | DBA | -0.27 | 0.68 | -0.37 | 0.13 | -2.1% | +0.8% | -35% | -62% | No effect |
| MR-16 Connors RSI | UNG | -0.22 | -0.27 | -0.15 | -0.35 | -4.0% | -25.1% | -60% | -99% | Clear and stable |
| MR-16 Connors RSI | BTC | 0.14 | 0.25 | 0.49 | 0.80 | +13.1% | +52.0% | -55% | -83% | No effect |
| MR-16 Connors RSI | ETH | 0.01 | 0.09 | 0.19 | 0.56 | +1.6% | +22.9% | -62% | -94% | No effect |
Sharpe ratio minus benchmark, full history · Benchmark = the buy-and-hold shown on the strategy pages.
The honest caveat
Effect is small and sensitive to costs and regime.
Descriptive statistics from historical data, before costs. Not a forecast and not investment advice.
Questions
After three down days in a row, the market bounces.
Not supported by the data: after “Three down closes in a row”, the average 3-day return was 0.0% against +0.1% for all days (medians over 10 markets); it was higher than usual in 3 of 10 markets and statistically clear, with at least 20 independent events and the same sign in both halves, in 0. The sign mostly points the other way, and the result cannot be told apart from chance.
How was this claim tested?
Event: three consecutive daily closes lower than the previous close. We record the return over the next three trading days (events at least 3 days apart) and compare it with the market's average 3-day return. The same pattern above the 200-day average is shown for contrast.
Related strategies on Quant Arena
Markets used
S&P 500 · Gold · Silver · Oil · Natural gas · Commodities · Agriculture · EUR/USD · Bitcoin · Ether