QQQ/SPY · > 2% below its 21-day MA · forward 21 sessions · recency window from 2021-01-01 · half-life 9mo · de-clustered ≥21d
Now: -2.1% vs the 21-day MA · 7th pctile (2021+) · 4th all-time
Close 0.92 · MA 0.94 · IN BREACH now · all-time range -5.9% … +7.3%
Forward returns by regime
Median forward return (% positive) at 2 / 5 / 8 / 11 / 15 / 21 sessions, conditioned on each market regime — and the recency-weighted read (the headline). Watch how the shape shifts as the window tightens toward now.
Full historyPost-CovidAI eraRallyRecency-wtdbaseline
Horizon
Full historyn=63
Post-Covidn=19
AI eran=7
Rallyn=7
Recency-wtdn=19
2 sessions
+0.2% 60%
+0.2% 63%
+0.1% 57%
+0.1% 57%
-0.0% 40%
5 sessions
+0.3% 60%
+0.0% 53%
+0.4% 71%
+0.4% 71%
+0.4% 82%
8 sessions
+0.2% 54%
-0.5% 32%
-0.2% 43%
-0.2% 43%
-0.2% 32%
11 sessions
+0.4% 59%
-0.3% 37%
+0.0% 57%
+0.0% 57%
+0.0% 52%
15 sessions
+0.2% 62%
+0.1% 53%
+0.1% 57%
+0.1% 57%
+0.1% 54%
21 sessions
+0.7% 60%
-0.3% 44%
+0.9% 67%
+0.9% 67%
+0.9% 75%
Regime delta: the recency-weighted 21-day median is +0.1 pts vs the full-history read — little regime change.
Headline is recency-weighted (faster 9-month half-life) over 19 post-2021 events. Regime windows: AI era from Apr 1 2023 · Rally from the Oct 30 2023 bottom. Tight regimes carry few events (see the n per column) — read them with the recency-weighted line, which keeps sample by fading rather than cutting.
Forward 21d · median
+0.9%
mean +0.5%
% positive
75%
19 events
vs baseline
+0.5 pts
base +0.4% / 58% pos
Range
-4% … +3%
25th -1.7% · 75th +1.3%
Path · max-gain
+1.3%
median MFE in window
Path · max-draw
-1.7%
median MAE in window
Every event · outcome over time
Each dot is one de-clustered event at its date; height is the forward 21-session return. Gold-ringed = post-2021 — if there are few gold rings, the signal is a pre-Covid-regime artifact.
By decade: 2000s 182010s 212020s 23
The event log
Event date
Stretch
Fwd 21d
Max gain
Max draw
2005-02-03
-2.3%
-0.7%
+0.8%
-2.1%
2005-04-19
-1.7%
+3.3%
+3.3%
-0.5%
2005-06-27
-1.8%
+3.9%
+4.1%
-0.5%
2006-02-06
-2.1%
-1.1%
+0.7%
-1.6%
2006-05-01
-1.7%
-4.2%
+0.3%
-4.2%
2006-05-31
-2.3%
+0.2%
+1.4%
-0.9%
2006-07-05
-1.8%
-3.7%
+0.0%
-4.5%
2006-12-19
-1.7%
-1.0%
+2.8%
-1.7%
2007-02-01
-1.8%
+0.9%
+2.4%
-0.4%
2007-08-16
-2.2%
+3.0%
+4.2%
+0.0%
2007-11-09
-2.2%
+0.7%
+2.0%
-1.6%
2008-01-07
-1.8%
-5.7%
+0.7%
-5.7%
2008-02-06
-3.4%
+0.8%
+2.4%
-1.0%
2008-07-22
-2.0%
+5.1%
+6.0%
+0.0%
2008-08-29
-1.9%
-6.8%
+0.0%
-6.8%
2008-09-30
-3.1%
+1.8%
+5.5%
-2.7%
2008-11-05
-2.1%
-1.0%
+1.9%
-2.8%
2009-05-07
-2.5%
+3.4%
+3.4%
-2.2%
2010-02-01
-2.0%
+2.4%
+2.4%
-0.3%
2011-03-18
-1.9%
+1.1%
+1.6%
+0.0%
2011-06-17
-1.8%
+4.7%
+4.7%
+0.0%
2011-10-27
-2.1%
-0.1%
+1.1%
-0.7%
2011-12-21
-2.2%
+2.3%
+2.6%
-0.1%
2012-04-24
-2.0%
+0.3%
+1.6%
-0.6%
2012-10-09
-2.1%
-1.6%
+0.3%
-1.9%
2013-01-24
-2.0%
-0.3%
+0.4%
-1.0%
2014-03-26
-2.1%
-2.0%
+0.3%
-2.2%
2015-08-21
-2.3%
+3.4%
+3.8%
+0.0%
2016-02-04
-1.7%
-1.3%
+0.7%
-2.1%
2016-04-22
-1.8%
-0.6%
+0.2%
-1.8%
2016-11-10
-2.9%
-1.5%
+1.0%
-1.5%
2017-06-12
-1.7%
+0.6%
+1.3%
-1.9%
2017-09-25
-2.0%
+0.6%
+1.7%
+0.0%
2017-12-01
-1.7%
+1.0%
+1.0%
-1.0%
2018-03-28
-2.2%
+0.5%
+1.7%
-0.6%
2018-07-30
-1.9%
+1.8%
+2.2%
+0.0%
2018-10-10
-2.3%
+0.8%
+2.6%
+0.0%
2018-11-12
-2.4%
+2.0%
+2.0%
-1.3%
2019-05-20
-1.8%
+0.7%
+0.7%
-2.2%
2020-04-09
-1.8%
+7.4%
+7.4%
+0.0%
2020-06-04
-2.0%
+7.7%
+7.7%
-1.0%
2020-07-24
-1.8%
+3.9%
+4.3%
+0.0%
2020-09-08
-1.7%
+1.0%
+3.0%
-0.8%
2020-11-09
-2.0%
+0.9%
+2.3%
-1.6%
2021-02-22
-2.3%
-2.4%
+0.0%
-5.7%
2021-05-04
-2.5%
-0.8%
+0.4%
-2.0%
2021-09-29
-1.9%
+1.4%
+1.4%
-0.5%
2021-12-16
-2.4%
-2.3%
+1.7%
-2.3%
2022-01-18
-2.2%
-1.8%
+0.0%
-2.9%
2022-02-23
-1.8%
+2.1%
+2.1%
-2.3%
2022-04-11
-2.0%
-4.1%
+0.8%
-4.1%
2022-05-11
-2.7%
-0.3%
+1.4%
-1.9%
2022-08-29
-1.8%
-0.3%
+0.2%
-1.3%
2022-10-11
-1.9%
-4.2%
+0.5%
-4.4%
2022-11-09
-2.6%
+1.9%
+3.2%
+0.0%
2022-12-22
-2.1%
+2.6%
+2.9%
-1.6%
2024-07-17
-2.2%
-0.8%
+0.5%
-3.1%
2024-09-06
-2.1%
+2.1%
+2.8%
+0.0%
2025-01-27
-1.8%
+0.9%
+2.9%
+0.0%
2025-02-27
-1.9%
-1.5%
+0.7%
-1.5%
2025-11-21
-1.8%
+0.8%
+2.2%
-0.1%
2026-02-04
-2.0%
+1.0%
+1.2%
-0.2%
Read it honestly. The recency-weighted read leans on 19 post-2021 events (half-life 9mo); the full-history column carries 63. The regime delta shows how far recent behaviour has drifted from the long base rate. De-clustered by a 21-session gap; overlapping windows remain. A context gauge, not a mechanical signal.
Computed live from adjusted-close history (yfinance). Forward return = close-to-close over 21 sessions from each event. MFE/MAE = max favorable / adverse excursion within the window. Educational research, not investment advice.