MA-Stretch Query · Recency-Weighted Event Study

AMD at the 200-day line

AMD  ·  > +51% above its 200-day MA  ·  forward 21 sessions  ·  recency window from 2021-01-01  ·  half-life 9mo  ·  de-clustered ≥42d
Now: +52.7% vs the 200-day MA  · 90th pctile (2021+) · 89th all-time
Close 476.15 · MA 311.82 · IN BREACH now · all-time range -67.4% … +128.3%

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.
-9.9%-0.5%+8.9%+18.3%+27.7%2d5d8d11d15d21d
Full historyPost-CovidAI eraRallyRecency-wtdbaseline
HorizonFull historyn=24Post-Covidn=6AI eran=5Rallyn=4Recency-wtdn=6
2 sessions-1.3% 42%-1.0% 50%+1.3% 60%-1.0% 50%+1.3% 62%
5 sessions-0.1% 50%-0.2% 50%+1.7% 60%+2.7% 75%+3.7% 58%
8 sessions-2.0% 38%+0.6% 50%-2.1% 40%+6.1% 50%+15.1% 54%
11 sessions-0.1% 46%+3.8% 67%+7.3% 80%+8.6% 75%+16.8% 95%
15 sessions-1.6% 42%+2.6% 50%-0.2% 40%+10.5% 50%+21.9% 54%
21 sessions-2.7% 46%-4.1% 33%-4.9% 40%+8.9% 50%+22.7% 54%
Regime delta: the recency-weighted 21-day median is +25.5 pts vs the full-history read — the recent regime is markedly stronger here.
Headline is recency-weighted (faster 9-month half-life) over 6 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
+22.7%
mean +16.5%
% positive
54%
6 events
vs baseline
+21.7 pts
base +1.0% / 53% pos
Range
-8% … +45%
25th -6.5% · 75th +16.2%
Path · max-gain
+8.5%
median MFE in window
Path · max-draw
-7.1%
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.
-50%-24%+2%+29%+55%median -2.7%2008201120142017202020232026each dot = one event · y = forward return
By decade: 2000s 4 2010s 12 2020s 8

The event log

Event dateStretchFwd 21dMax gainMax draw
2005-12-27+52.7%+27.7%+29.5%-2.1%
2006-02-28+54.2%-11.3%+6.9%-12.2%
2009-09-08+61.6%+6.7%+17.1%+0.0%
2009-11-12+62.2%+34.0%+34.4%-0.8%
2010-01-14+71.1%-10.4%+0.1%-20.1%
2010-03-17+56.8%+1.4%+5.1%-6.7%
2013-05-15+52.5%-10.0%+0.0%-12.6%
2013-07-16+52.5%-13.8%+4.7%-18.1%
2016-04-22+83.1%+1.3%+1.3%-13.5%
2016-06-22+89.1%+16.3%+16.3%-6.0%
2016-08-22+116.4%-17.0%+1.2%-24.3%
2016-10-20+57.7%+25.1%+25.1%-9.5%
2016-12-20+100.6%-13.9%+4.9%-15.3%
2017-02-22+88.4%-3.4%+6.4%-8.8%
2018-07-27+51.9%+33.4%+33.4%-3.2%
2018-09-26+111.6%-40.1%+1.2%-40.1%
2020-01-02+53.0%-2.2%+5.3%-4.3%
2020-07-29+58.6%+10.1%+14.0%+0.0%
2021-11-08+60.9%-3.3%+7.8%-7.4%
2023-05-25+51.4%-8.3%+7.3%-10.7%
2024-01-19+54.5%-4.9%+3.5%-4.9%
2025-10-06+58.6%+22.7%+29.8%+0.0%
2026-04-24+68.4%+44.9%+44.9%-7.1%
2026-06-25+98.2%-7.1%+9.1%-7.1%
Read it honestly. The recency-weighted read leans on 6 post-2021 events (half-life 9mo); the full-history column carries 24. 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.