1 · Self-audit — is the brain alive & honest right now?

The worker checks its own vitals every minute (and a GitHub monitor emails on failure). Green = healthy. This is the same audit_pass the uptime alarm watches.
Checking…
Loading…

2 · Does it beat a coin flip?

Out-of-sample, walk-forward (train on the past, test on the held-out future). A directional call is "right" if price moved its way over 5 trading days. p-value < 0.05 means the edge is unlikely to be luck. Full breakdown on the Edge Scorecard.
Walk-forward accuracy
Statistically significant?
Sample size
Brier skill vs random
Trained on

3 · Where the edge actually lives (the brain's learned map)

Hit-rate split by how confident the brain was. This is what it has learned about its own edge: the bands that beat 52% are signal; the rest is noise we filter out. Two views — the backtest (thousands of out-of-sample calls, available now) and the live forward record (accrues at 5 days).
Backtest — hit-rate by conviction band
Loading…
Live forward record (graded at 5 days)
Loading…

4 · The live track record of what we broadcast

Every Pick of the Day and Alpha call is logged with its entry price and graded against the real 5-day move — no cherry-picking. Records start empty and fill in over weeks.
🎯 Pick of the Day record
Loading…
⭐ Alpha record
Loading…

5 · Noise control — what we filter, and how much

The brain scores the whole universe but we only broadcast the calls that pass your constraints. The brain still trains on everything — this is purely what gets surfaced. Edit the constraints →
Universe scored
Passed constraints
Filtered as noise
Min conviction

6 · Why each call makes sense (today)

For every broadcast call: the brain's probability, today's move, relative volume, and the plain-English reason. A call is a thesis you can read, not a black box.
Loading…

7 · The methodology, end to end

Exactly how a price becomes a tracked call. No magic. (The browser Trade-of-the-Day desk uses a separate, fully-public 8-factor score — see Scoring Methodology.)
1
Data. The 24/7 Cloudflare worker pulls real prices from Finnhub (live) and Yahoo (resolution, ~15-min delayed), real SEC Form-4 insider filings, and real market news — no fabricated numbers. It rotates through a ~72-symbol universe every few minutes, market hours only.
2
Features. Each symbol becomes a 22-dimensional vector — momentum, volatility, relative volume, distance from moving averages, RSI, day-of-session (ET-anchored), and more — computed from a rolling per-symbol bar history kept in KV so live captures match the training set exactly.
3
Model. Logistic regression with an Adam optimizer and L2 weight decay (0.025), trained on the most recent ~120 days so it shares the current regime. Output is P(up over the next 5 trading days, by at least 1%).
4
Calibration. A Platt layer maps raw scores to honest probabilities — but only if it passes a guard (slope a ≥ 0.2); a noisy fit that would invert the sign is rejected and we fall back to raw. The self-audit above shows the live Platt slope.
5
Selection. Conviction = how far P(up) is from a coin flip. The single highest-conviction passing call is the Pick of the Day; the most-bullish passing name is the Best Long; the top passing leans are Alpha. Your constraints (section 5) decide what "passing" means.
6
Learning loop. Every call is journaled with its entry price, then graded against the real move 5 trading days later. Those graded outcomes feed back into training — so the brain improves from real results, and section 3 shows which confidence bands actually earned their keep.
7
Honesty. Backtest ≠ live. The walk-forward number (section 2) is the out-of-sample estimate; the live record (section 4) is the real forward test and starts empty. We never present one as the other.
Educational research tool, not financial advice. A statistically-significant edge is still a small edge — position size accordingly and assume drawdowns.