| Date | 15d Trigger Move | PSCE +10d | PSCE +20d | Ratio +10d | Ratio +20d | Ratio +40d |
|---|---|---|---|---|---|---|
| 2026-06-09 | -11.2% | -6.4% | -5.3% | -9.6% | -9.2% | -8.2% |
| 2026-04-15 | -11.3% | +9.0% | +5.4% | +7.0% | +3.1% | -5.2% |
| 2025-04-10 | -12.7% | +8.8% | +13.4% | +2.7% | +3.1% | +3.5% |
| 2025-02-06 | -13.3% | -3.4% | -14.7% | +1.6% | -6.0% | -10.4% |
| 2023-11-08 | -11.2% | +2.6% | -3.3% | -1.9% | -11.1% | -13.3% |
| 2022-12-07 | -12.5% | +3.7% | +1.2% | +6.0% | +1.7% | -0.5% |
| 2022-09-26 | -12.3% | +21.9% | +30.5% | +18.6% | +22.1% | +20.5% |
| 2022-06-22 | -11.1% | -6.2% | -5.0% | -9.3% | -12.0% | -3.7% |
| 2021-11-19 | -13.0% | -1.6% | -8.6% | +1.8% | -1.8% | +15.4% |
| 2021-07-16 | -12.9% | -0.6% | -6.9% | -3.8% | -10.6% | -0.9% |
| 2021-03-26 | -11.6% | -8.2% | -7.3% | -8.5% | -9.1% | +7.3% |
| 2020-09-02 | -12.8% | -2.9% | -14.5% | +1.8% | -9.6% | -8.1% |
| 2020-06-26 | -17.5% | -0.3% | +12.5% | -3.1% | +2.7% | +0.5% |
| 2020-01-21 | -12.6% | -10.8% | -10.7% | -8.3% | -9.6% | -43.4% |
| 2019-10-01 | -11.7% | -1.7% | +0.3% | -3.1% | -5.1% | -13.3% |
| 2019-07-18 | -11.2% | -2.3% | -16.6% | -2.9% | -12.6% | -6.1% |
| 2019-05-02 | -11.8% | +1.5% | -16.9% | +3.5% | -9.5% | -9.9% |
| 2018-10-29 | -14.1% | -0.7% | -8.7% | -3.5% | -9.8% | -24.3% |
| 2018-08-20 | -11.4% | +0.8% | -1.1% | -0.8% | -1.3% | +6.7% |
| 2018-06-12 | -11.9% | +1.2% | +2.8% | +2.1% | +2.5% | -4.8% |
| 2017-04-27 | -11.1% | -5.1% | -12.0% | -3.3% | -9.5% | -20.7% |
| 2017-03-08 | -14.6% | -3.1% | -0.2% | -1.9% | +1.2% | -17.8% |
| 2016-07-25 | -12.5% | +7.8% | +11.2% | +6.8% | +9.4% | +6.5% |
| 2015-12-07 | -12.7% | -9.1% | -11.2% | -6.1% | -5.9% | -8.9% |
| 2015-09-23 | -11.3% | +15.6% | +9.5% | +12.9% | +7.2% | +4.6% |
| 2015-07-01 | -12.8% | -7.4% | -17.0% | -8.1% | -15.5% | -22.1% |
| 2014-11-28 | -14.4% | -14.8% | -1.4% | -13.2% | -5.2% | -14.5% |
| 2014-10-08 | -11.7% | -9.7% | -9.3% | -10.1% | -15.4% | -31.3% |
| 2011-10-06 | -11.8% | +11.5% | +23.9% | +8.1% | +11.4% | +13.1% |
| 2011-08-09 | -11.3% | -1.9% | +4.3% | -0.5% | +1.8% | -9.4% |
| 2011-05-13 | -11.3% | +5.7% | -3.8% | +5.8% | +3.0% | +7.4% |
| Horizon | N | PSCE Mean | PSCE Median | PSCE Win% | IJR Median | IJR Win% | Ratio Mean | Ratio Median | P(Lower) | Edge vs Base |
|---|---|---|---|---|---|---|---|---|---|---|
| Day 1 | 31 | -0.5% | -0.4% | 39% | -0.6% | 35% | -0.0% | +0.2% | 42% | +0.0% |
| Day 3 | 31 | +0.4% | +1.4% | 58% | +0.4% | 61% | +0.0% | -0.1% | 52% | +0.1% |
| Day 5 | 31 | -0.1% | +0.0% | 48% | +0.9% | 61% | -0.6% | -1.3% | 55% | -0.4% |
| Day 10 | 31 | -0.2% | -1.6% | 39% | +0.7% | 58% | -0.6% | -1.9% | 58% | -0.2% |
| Day 15 | 31 | -0.4% | -2.0% | 39% | +0.7% | 65% | -1.5% | -2.4% | 65% | -1.0% |
| Day 20 | 31 | -1.9% | -3.8% | 35% | +1.7% | 61% | -3.2% | -5.2% | 61% | -2.4% |
| Day 30 | 31 | -1.2% | -1.8% | 42% | +1.2% | 68% | -3.5% | -3.6% | 61% | -2.3% |
| Day 40 | 31 | -4.4% | -2.5% | 45% | +1.8% | 65% | -6.2% | -6.1% | 68% | -4.7% |
| Day 50 | 31 | -4.3% | -1.6% | 42% | +4.6% | 65% | -6.2% | -7.3% | 68% | -4.3% |
Two mechanical scale-out overlays on the same trigger. Enter 100% within two days of the signal, then manage the position in tranches — 20% / 30% / 30%, with the final 20% always exiting at the end of the study window.
Where the deviations come from. For every completed historical precedent we measure how far the trade travelled in your favor (its Maximum Favorable Excursion). The three profit deviations span that favorable-outcome distribution — a low, median and high reading of how far these trades historically ran. (For thin samples these are the literal weakest / typical / strongest outcomes; for larger samples they are robust quartiles, so one outlier can't set the rungs.) The three loss deviations are the exact mirror of the profit deviations — the same levels flipped below the entry, so a +4% first take-profit pairs with a −4% first stop. It's a symmetric bracket. A rung only fills if price actually trades there; unfilled tranches exit with the balance at the window close. Mirrored for shorts (the profit-taker buys the name back lower; the loss-taker covers higher).
⚠️ Small-sample caveat. With only a handful of completed precedents these levels are illustrative, in-sample anchors — a disciplined way to turn "what this pattern historically did" into entry, take-profit and stop levels, not a statistically validated system. The live bracket is the payoff, not the backtest averages.
| Rung | Return | Price | Action | Live status |
|---|---|---|---|---|
| Profit · 3rd dev | +17.4% | 0.355 | cover 30% | pending |
| Profit · 2nd dev | +12.4% | 0.377 | cover 30% | ● filled |
| Profit · 1st dev | +5.9% | 0.404 | cover 20% | ● filled |
| ● Entry | 0.0% | 0.43 | short 100% | Day 50 |
| Loss · 1st dev | -5.9% | 0.455 | cut 20% | pending |
| Loss · 2nd dev | -12.4% | 0.483 | cut 30% | pending |
| Loss · 3rd dev | -17.4% | 0.504 | cut 30% | pending |
| Signal date | Peak (MFE) | Trough (MAE) | Buy & hold | Profit-taker | P-devs | Loss-taker | L-devs |
|---|---|---|---|---|---|---|---|
| 2011-05-13 | +1.0% | -15.8% | -15.8% | -15.8% | ○○○ | -12.8% | ●●○ |
| 2011-08-09 | +11.7% | -5.5% | +3.7% | +4.1% | ●○○ | +3.7% | ○○○ |
| 2011-10-06 | -0.0% | -14.7% | -6.1% | -6.1% | ○○○ | -7.9% | ●●○ |
| 2014-10-08 | +36.5% | -0.0% | +27.1% | +15.5% | ●●● | +27.1% | ○○○ |
| 2014-11-28 | +13.0% | -1.0% | +6.1% | +8.0% | ●●○ | +6.1% | ○○○ |
| 2015-07-01 | +27.1% | -0.0% | +24.8% | +15.1% | ●●● | +24.8% | ○○○ |
| 2015-09-23 | +5.2% | -16.6% | +5.2% | +5.2% | ○○○ | -2.3% | ●●○ |
| 2015-12-07 | +23.1% | -3.4% | +18.6% | +13.8% | ●●● | +18.6% | ○○○ |
| 2016-07-25 | +4.3% | -17.0% | -17.0% | -17.0% | ○○○ | -13.4% | ●●○ |
| 2017-03-08 | +16.7% | -5.7% | +10.8% | +10.3% | ●●○ | +10.8% | ○○○ |
| 2017-04-27 | +21.3% | -0.0% | +18.9% | +13.9% | ●●● | +18.9% | ○○○ |
| 2018-06-12 | +10.4% | -4.5% | +6.1% | +6.0% | ●○○ | +6.1% | ○○○ |
| 2018-08-20 | +8.7% | -8.3% | +7.9% | +7.5% | ●○○ | +5.2% | ●○○ |
| 2018-10-29 | +25.5% | -2.2% | +17.1% | +13.5% | ●●● | +17.1% | ○○○ |
| 2019-05-02 | +17.6% | -3.4% | +16.9% | +13.5% | ●●● | +16.9% | ○○○ |
| 2019-07-18 | +14.2% | -2.3% | +13.8% | +11.8% | ●●○ | +13.8% | ○○○ |
| 2019-10-01 | +13.3% | -0.9% | +4.9% | +7.4% | ●●○ | +4.9% | ○○○ |
| 2020-01-21 | +49.3% | -0.0% | +44.5% | +19.0% | ●●● | +44.5% | ○○○ |
| 2020-06-26 | +8.0% | -6.1% | +6.7% | +6.5% | ●○○ | +4.2% | ●○○ |
| 2020-09-02 | +15.2% | -1.4% | +6.8% | +8.3% | ●●○ | +6.8% | ○○○ |
| 2021-03-26 | +10.2% | -20.5% | -19.3% | -14.2% | ●○○ | -13.9% | ●●● |
| 2021-07-16 | +15.2% | -11.0% | -11.0% | -0.6% | ●●○ | -9.9% | ●○○ |
| 2021-11-19 | +3.7% | -19.6% | -19.3% | -19.3% | ○○○ | -14.0% | ●●● |
| 2022-06-22 | +8.4% | -12.2% | -10.1% | -6.9% | ●○○ | -9.2% | ●○○ |
| 2022-09-26 | -0.0% | -27.8% | -7.0% | -7.0% | ○○○ | -11.5% | ●●● |
| 2022-12-07 | +2.7% | -10.8% | +2.7% | +2.7% | ○○○ | +1.0% | ●○○ |
| 2023-11-08 | +17.9% | -1.0% | +15.1% | +13.1% | ●●● | +15.1% | ○○○ |
| 2025-02-06 | +11.5% | -3.7% | +8.5% | +8.0% | ●○○ | +8.5% | ○○○ |
| 2025-04-10 | +3.3% | -10.4% | -1.0% | -1.0% | ○○○ | -1.9% | ●○○ |
| 2026-04-15 | +15.5% | -7.8% | +15.5% | +12.7% | ●●○ | +11.3% | ●○○ |
The study isolates relative-strength washouts in small-cap energy: sessions where the PSCE/IJR ratio (Invesco S&P SmallCap Energy vs iShares Core S&P Small-Cap) has fallen more than 11% over the trailing 15 sessions. Clusters are deduplicated — only the first qualifying day of each episode counts, with a minimum 20-session gap. Over the full PSCE lifetime (April 2010 – present, 4,126 sessions), this has occurred 31 times, most recently on Jun 09, 2026 — the live trigger. The current episode took the ratio from a May 19 local peak of 0.479 to roughly 0.419, a −12.6% relative-strength collapse.
This is the mirror image of the META/QQQ result. Where the mega-cap divergence mean-reverts, the small-cap energy washout continues. Twenty sessions after the trigger, the ratio's median forward return is -5.2% against an unconditional baseline of -0.7% — the spread keeps falling at roughly 7x its normal drift rate. The probability the ratio is lower at day 20 is 61%, rising to 68% at day 40.
The components tell the same story: PSCE itself posts a 20-day median of -3.8% (win rate 35%) while IJR holds firm at +1.7%. The damage is concentrated in the energy leg — the benchmark is not the problem. The first ~5 sessions often produce a dead-cat bounce (day-5 PSCE win rate 48%), which historically has been the selling opportunity, not the bottom.
The ratio chart shows a secular downtrend punctuated by violent countertrend rallies (2016, 2020–2022). Small-cap energy is a high-beta expression of crude fundamentals with weak balance sheets: when the relative tape breaks this hard, it is usually because the underlying commodity or credit picture deteriorated — conditions that persist for weeks, not days. Momentum in the spread is informationally efficient: the washout is the news. Buying the dip fights both the secular trend and the freshly confirmed local trend.
The quant expression follows the data: short the spread, do not fade it.
With 31 events the sample is workable but clustered — many triggers occur inside the same macro regimes (2014–16 oil bust, 2020 COVID), so the effective number of independent observations is lower than 31. Standard errors on the continuation probabilities are roughly ±18 pct pts. PSCE is a thin ETF (~$80M AUM); use limit orders and check borrow availability on the short leg. Past conditional probabilities do not guarantee future returns.