jev-trader
Trade Arena proof of concept comparing trading strategies on 13 US stock funds, grown out of a pre-registered Jev study.
Overview
jev-trader (Trade Arena) is a site where you pick two well-known trading strategies and see, in plain words, what $1,000 became in each on the same 13 US stock funds. It is a proof of concept that grew out of a finished study of Jev, TypeSafe's decision model, against technical-analysis rules.
A research directory computes the results, a `jt export` command writes the Jev study's data, `jt arena` writes the arena data, and a static Next.js site reads only those files. The site currently lands on the Jev study's pre-registered holdout result, with the Pick two prototype on a preview page; the plan is to move the study to its own page and make the arena the home page. Product, implementation and design-decision documents are included and still changing. The README itself does not state the study's outcome.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Should the strategy buy, sell or hold this fund? (Jev arm of the study)(inferred) | — |
How it works
- Research code computes strategy results on 13 funds
- Export study and arena data files
- Static site reads the files and presents plain-language comparisons
Strengths
- Pre-registered holdout design for the Jev study
- Static site reads only exported data, keeping numbers consistent
Caveats
- Proof of concept; plans still changing
- Study result not stated in the README
At a glance
- Type
- Project
- Status
- prototype
- Language
- Python
- Runtime
- CLI research + static Next.js site
- Has code
- yes
Use cases and integrations
How curated lists describe it
Can jev predict the stock market? Let.
wh000wh000/awesome-jev-live
Frequently asked questions
What is Trade Arena?
A site comparing what $1,000 became under two chosen strategies on the same 13 US stock funds.
How does Jev fit in?
The project grew out of a pre-registered study of Jev against technical-analysis rules, shown on the site.
Is it finished?
No, it is a proof of concept with milestones still planned.
Related projects
PyTorch factor-research stack with 213 factors and cost-aware backtests, plus a preview typed portfolio-decision adapter.
ProjectbetaPythonReproducible test of viral Jev HFT claims: direction not calibrated, never under 100 ms, and no edge after trading fees.
ServiceresearchPythonPoint-in-time O'Neil momentum backtest where Jev picks entries and judges exits, with a jackknife showing no resolvable edge over rules.
ProjectresearchPythonDecision Lab: sixty finance demos asking Jev, OpenAI Decisions and twelve other decision models the same questions, graded.
ProjectresearchJavaScriptBacktest harness that tests Jev as a BUY/SELL/HOLD trader on NQ level-10 order-book data, with replay, viewer and paper-trading modes.
ProjectresearchPythonStudy of whether Jev's reading of financial news predicts stock returns: strong same-day signal, no tradeable alpha.
BenchmarkresearchGo