Sample · Stock Model Assessment
NVDA
210.96
▼ −0.59% · close 2026-10-23
Close sits inside the pivot zone. A close below 194.30 breaks the structure.
US equities · quantitative research
US stock research you can check.
Every reading carries a data date, a fingerprint and the condition that would prove it wrong. Read it here, or ask from ChatGPT, Claude, Grok or Cursor.
Free sample, no sign-in.
One page, three depths.
The sentence takes five seconds. Open the numbers for proof. Open the method for the math.
Read the conclusion and the condition that would prove it wrong.
Check the data date, the sample and the comparison.
Open the method for definitions and limits.
What arrives, and when.
Every report shows its data date. If your plan cannot see today's edition, the answer says which date it used.
Stock Model Assessment
One ticker · 20 trading days · three layers
Pre-market · Intraday · Close
Three market briefs a trading day
Model Desk Daily
Daily readings of the four models
House View
House view, Mondays
Company Research · Thesis & Counterarguments
The case for, the case against, what decides it
Company Research · Fundamentals
Filings-based financial read
Event study
Abnormal returns around a dated event
Report history
Every past edition, by date and type
Bench · coming soon
Open Bench
Check the idea before you trust it.
Each result shows whether it could be luck, what is left after costs, and when it stopped working before.
Risk exposure
How much of a move is just the market
Model screener
What changed today, where models disagree
Factor evaluation
Every trial counted, failures included
Portfolio check
Hidden concentration in your holdings
Idea check
How close an idea is to a coin flip
Market regime
Is today's market unusual?
Built so you can audit us.
fingerprint
A content hash and timestamp on every report. You can prove what it said, and when.
failures kept
Tests that did not pass stay on record and in the denominator.
luck check
Placebo draws and multiple-testing corrections next to each result.
invalidation
A mechanical condition that, if met, means the reading is wrong.