Can AI pick tickers and auto-trade them for a profit?
I’m going to find out. Rules pick the names, the machine buys and sells, and I write up the seats that broke.
By date, ±%
Closed names only. No dollars. Only days we actually logged.
- unrealized+0.15%10-seat swing+1.12%sleeverealized+11.85%DE+5.61%VGT−10.59%PANW
Momentum SEPA swing · Tripod sleeve
DE · Target exit +11.85% · VGT · Sleeve exit +5.61% · PANW · Stop −10.59% · DRAM · Rule exit −19.40%
- Unrealized+2.74%Breakout+2.63%MomentumRealized—
Breakout scanner · Momentum scanner · Small-cap paper
—
- Unrealized—Realized−11.35%
—
PYPL · ATR stop −11.35%
- Unrealized—Realized—
—
—
- Unrealized—Realized—
Breakout scanner · Momentum scanner
—
- Unrealized—Realized−4.38%
—
PANW · lost the 20-day line −4.38% · ANET · lost the 20-day line
Recent
all posts →- backtest
The 4R Target Illusion vs 2.0R Split Trailing
With a 4R target, the actual reach rate was only 5.2%. Why taking half at 2.0R and trailing the rest works.
- backtest
Why Buying in Daytime Sessions Hurts: The Cost of Gaps
Buying early in daytime ATS looks convenient, but overnight gap-downs straight through stops erode strategy edge.
- log
Why I ran the momentum paper book again
On August 27 I started the momentum paper book over, dropping names weaker than their sector.
- log
August 30 paper book, pluses and minuses
Open seats stay blurred, ±% only. Closed names: PYPL, PANW, ANET.
- log
Why I opened a separate paper book for smaller names
A 500-name paper book opened August 28. No auto-buy. No rows yet.
- log
Why I ran the breakout paper book again
On August 27 I started the breakout paper book over with a filter that drops names weaker than their sector.
- backtest
Does the breakout rule break in the 2022 bear?
The full sample looks like a pass. 2022’s bear and 2023’s chop still lose money.
- backtest
Drop names weaker than their sector — does it help?
Dropping weaker-than-sector names cuts trades ~14% and lifts Sharpe 0.35→0.40. Bears still lose.