MA-Stretch Query · Recency-Weighted Event Study

XLK/XLF at the 65-day line

XLK/XLF  ·  > 6% below its 65-day MA  ·  forward 21 sessions  ·  recency window from 2021-01-01  ·  half-life 9mo  ·  de-clustered ≥42d
Now: -8.6% vs the 65-day MA  · 3th pctile (2021+) · 2th all-time
Close 3.08 · MA 3.37 · IN BREACH now · all-time range -24.3% … +43.9%

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.
-1.2%+1.0%+3.2%+5.4%+7.6%2d5d8d11d15d21d
Full historyPost-CovidAI eraRallyRecency-wtdbaseline
HorizonFull historyn=25Post-Covidn=12AI eran=6Rallyn=6Recency-wtdn=12
2 sessions+0.7% 72%+0.8% 83%+1.8% 100%+1.8% 100%+0.5% 99%
5 sessions+2.1% 88%+3.3% 100%+3.0% 100%+3.0% 100%+2.2% 100%
8 sessions+1.3% 84%+3.9% 83%+2.6% 83%+2.6% 83%+0.9% 54%
11 sessions+1.1% 76%+3.2% 83%+4.4% 83%+4.4% 83%+0.5% 54%
15 sessions+2.7% 75%+3.1% 100%+6.0% 100%+6.0% 100%+6.0% 100%
21 sessions+2.1% 71%+6.5% 82%+6.5% 80%+6.5% 80%+6.5% 75%
Regime delta: the recency-weighted 21-day median is +4.4 pts vs the full-history read — the recent regime is markedly stronger here.
Headline is recency-weighted (faster 9-month half-life) over 12 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
+6.5%
mean +4.3%
% positive
75%
12 events
vs baseline
+5.9 pts
base +0.6% / 56% pos
Range
-3% … +9%
25th +2.1% · 75th +7.3%
Path · max-gain
+7.2%
median MFE in window
Path · max-draw
-0.6%
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.
-14%-7%-1%+5%+12%median +2.1%2008201120142017202020232026each dot = one event · y = forward return
By decade: 2000s 7 2010s 5 2020s 12

The event log

Event dateStretchFwd 21dMax gainMax draw
2006-05-16-6.6%+0.8%+1.5%-2.8%
2006-07-20-6.5%+5.0%+5.0%-0.5%
2008-01-23-7.1%+0.9%+5.3%-2.8%
2008-09-03-7.4%-5.5%+7.2%-8.5%
2009-04-09-9.3%-11.2%+5.7%-16.6%
2009-06-10-6.5%+5.8%+5.8%-1.3%
2009-08-10-7.9%+2.0%+3.4%-1.2%
2012-10-18-7.1%-1.5%+0.3%-1.5%
2013-01-25-6.2%+1.2%+1.4%-0.4%
2016-11-10-10.6%-4.4%+0.6%-5.8%
2017-07-03-6.2%+4.8%+6.9%+0.0%
2018-11-19-6.7%+7.0%+10.4%+0.0%
2020-11-10-8.0%-0.7%+3.6%-2.6%
2021-01-12-8.9%+7.3%+10.1%-0.6%
2021-03-15-9.6%+4.0%+5.8%-2.3%
2021-05-13-9.5%+8.1%+8.1%-0.3%
2022-01-25-6.9%+2.1%+5.3%-2.5%
2022-09-15-6.4%-3.2%+3.9%-3.2%
2023-01-05-6.8%+9.3%+11.4%+0.0%
2024-04-19-6.9%+7.2%+7.2%-0.3%
2024-07-30-8.6%+2.1%+7.6%-0.8%
2024-11-27-6.4%+7.2%+8.5%+0.0%
2025-02-03-6.4%-0.7%+5.8%-4.9%
2025-04-03-9.4%+6.5%+6.8%-0.3%
Read it honestly. The recency-weighted read leans on 12 post-2021 events (half-life 9mo); the full-history column carries 25. The regime delta shows how far recent behaviour has drifted from the long base rate. De-clustered by a 42-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.