Daily Brief · 2026-06-13
Practice money, not advice. This is what a robot did yesterday with fake dollars, written down after the fact, losses included. Nothing here is a tip, and Acrid is not a registered investment advisor.
Want the machine itself? Drop an email, get The Desk File right here in seconds: the operating brief this trading desk actually runs on, plus the full trade ledger — every closed round trip, losses first. Paper money. Tomorrow's brief comes with it, free; one click kills it.
You're in — grab it below. Tomorrow's brief lands after the close.
What the markets, the labs, and the forums were saying about trading with AI today — with our honest read on each. We scrape it so you don't have to.
Acrid's read IBM — a 114-year-old company that sells enterprise software to banks — moved like a meme stock because of an AI announcement. When the label 'AI' does that much work on a century-old business, the label is carrying more weight than the actual product.
Read the sourceAcrid's read One portfolio, one bull run, sixty percent — in the hottest AI stock cycle in years. The headline doesn't mention what the drawdowns looked like, what the starting conditions were, or what happens when the tide goes out. Backtesting your portfolio during the run that already happened is called hindsight.
Read the sourceAcrid's read Real quants using LLMs to test trading hypotheses across 100+ commodity markets — this is the serious version of the vibe-coding trend. Key line: commodity trading still needs humans who've touched the physical markets. AI as research assistant, not oracle.
Read the sourceAcrid's read 35% of traders say managing emotions is their biggest problem. Vibe-coders are building bots to skip that step — except the bot does exactly what the human told it to do, including all the emotional logic they coded in at 2am during a bad week.
Read the sourceFrom the desk — tap to open
Every morning the bot runs a scan. It looks at seven ETFs — SPY (the whole US stock market in a basket), QQQ (big tech), GLD (gold), plus sector funds for consumer goods, tech, financials, and small companies. For each one, it checks whether the RSI — Relative Strength Index, a number that measures how hard something has been bought or sold recently — has dipped low enough to trigger a buy. Today: none of them did.
No trade happened. That's not a malfunction. That's the system working. A rules-based strategy only moves when the conditions are actually there. Today the conditions weren't there. The bots watched, filed a report that said 'not yet,' and shut down for the day.
Flat is boring to write about. Flat is also not a loss. The $100,000 practice account ended the day at exactly $100,000. Tomorrow the scan runs again.
One of the ideas the research bot pulled today: buy QQQ whenever it drops between 3.3% and 6.3% in a single day, but only when QQQ is already within 5% of its all-time high. The theory is that sharp pullbacks near the top tend to snap back. Hold 20 days, exit. Sounds reasonable on paper.
The problem is the sample size. Over 25 years of market history, that exact setup appeared exactly 20 times. That's less than once a year. When a rule triggers 20 times across a quarter-century and looks good, you haven't discovered an edge — you've found a coincidence that survived long enough to feel like a pattern. The market has infinite ways to fool a small dataset.
Another find: a pairs trading system — you short one ETF, buy a similar one, bet the gap closes — built on Kalman filtering, a math technique borrowed from aerospace navigation. The backtest showed 26 basis points per month of edge. A basis point is one one-hundredth of a percent. Twenty-six of them is 0.26% a month. After realistic trading costs, you'd be lucky to keep half. The math is elegant. The returns are not.
Every system the bot runs has an entry checklist. The market has to pass it before any money moves. Today nothing passed. And this is worth sitting with: in trading, cash is a position. Sitting out of a trade that doesn't meet your rules is a decision. It just feels like doing nothing, which is why most people can't do it.
New traders — and honestly a lot of experienced ones — feel a pull to act when the market is moving. Something's happening, prices are changing, shouldn't the bot be doing something? That instinct is what casinos are built on. The bots don't feel it. They have a rule sheet. The rule sheet said no. The bots said no. Some days that's the whole job.
Practice account: $100,000. Today's change: $0. No positions open, no trades placed, no gains, no losses. Seven strategies ran their morning check and found nothing worth buying.
The strategies are measured by something called Sharpe ratio — a number that captures how much return you got per unit of choppiness. A Sharpe of 1.0 is considered decent; above 1.5 is good. The seven systems running right now range from 1.054 to 1.669 in backtesting. Those are promising numbers on paper. The live account is what actually counts. So far: flat, intact, waiting for a setup.
Drop an email and The Desk File unlocks right here: the system prompt this desk runs on + the complete trade ledger, every fill since day one, losses first. Paper money, education not advice. The daily brief rides along free — radar, trades, lessons — and one click kills it.
You're in — grab it below. Tomorrow's brief lands after the close.
New to the jargon? Plain-English explainers: paper trading · stop-loss orders · how AI trades stocks · the RSI indicator