US Outcome & Track Record Decided by: AI 🤖

AI · US Semiconductors — Track record

Every elapsed decision is graded against what prices actually did — hit-rates and errors per signal family, plus the cost of the decisions this book chose not to take.

All books

48 graded 0 pending 1W calibration -0.23 How far along the 1-week path a decision typically gets: the fraction of the decided move realized after one week, measured across this book's graded decisions. Feeds a bounded conviction adjustment back into future decisions.

Scorecard

Hit-rates per signal family.

Signal family	Dir	n	Hit 1W	Hit 4W	Hit 12W	Med err 1W	Stop-first	Bias	Status
NoTrade	sell	18	18%	10%	—	—	0%	-6	● active
FailedBreakdown_Up_20D	sell	14	69%	—	—	—	0%	+4	● active
FailedBreakdown_Up_20D	buy	6	40%	0%	—	-7.5%	50%	-2	● active
RallySell_BearTrend	sell	5	40%	25%	—	—	0%	-2	● active
FailedBreakout_Down_20D	sell	2	0%	100%	—	—	0%	0	○ building sample (2/5)
AnticipationUp_VCP	buy	1	0%	—	—	-8.7%	0%	0	○ building sample (1/5)
AnticipationUp_VCP	sell	1	100%	—	—	—	0%	0	○ building sample (1/5)
ReversalUp_Oversold	sell	1	0%	0%	—	—	0%	0	○ building sample (1/5)

Counterfactuals Counterfactuals grade decisions that were NOT executed — expired queues, no-chase rules, untaken optional trims — against what the market then did. They are kept strictly separate from the calibration cells above.

What the decisions we didn't take would have done.

Optional trims not taken — 16 cases , median would-have 1W move -5.1% → verdict: cost
Queue expiry (order lapsed unfilled) — 2 cases , median would-have 1W move +2.9% → verdict: cost
Anticipate slot (pre-breakout buy-stop) — 9 cases → verdict: n/a
No-chase queue (queued a pullback instead of chasing) — 3 cases , median would-have 1W move +2.9% → verdict: cost

Counterfactuals are advisory evidence for the book's learnings reviews — never blended into the calibration cells.