In one test of Jev version 1.13.0, it did not predict the direction of the next BTCUSDT five-minute candle better than chance. On 1,500 unseen candles, Jev’s AUC was 0.486 with numeric indicators and 0.494 with text descriptions. A gradient-boosting model using the same selected indicators scored 0.522, but its confidence interval also included 0.5. The study’s more striking finding was directional: Jev’s interpretations tracked familiar momentum signals while the tested price data showed a slight tendency to reverse after stretched moves.
What “predict Bitcoin” meant in this test
The experiment addressed a tightly defined question: after a BTCUSDT five-minute candle closed, would the next candle close above its own open? That is a short-horizon direction classification task, not a forecast of Bitcoin’s price target, a test of long-term investing, or a complete trading strategy.
As an Amazon Associate I earn from qualifying purchases.
The findings were reported by the author using the handle truongxx in an article dated September 27, 2026. They concern Jev version jev-1.13.0, Binance BTCUSDT data, and price-derived indicators. They do not establish how another Jev version would perform, or how it would handle news, order-book data, or other markets.
Recommended Free Tools
How the experiment was set up
The author says candle data came from Binance’s public API. The reported training span was January 1, 2022, through December 31, 2024, with 314,552 candles. The validation period was July through December 2024 and informed choices. The final test span ran from January 1, 2025, through September 24, 2026, comprising 181,290 candles.
#1 Best Overall
- BITCOIN EXCLUSIVE, PHONE VERIFICATION: Bitkey is designed from the ground up exclusively for bitcoin — a dedicated hardware wallet for secure bitcoin storage. Approve transactions with a tap using your phone and NFC. No device screen is required.
- SELF-CUSTODY, NO EXCHANGE OR CUSTODIAN REQUIRED: You hold two of the three keys in the Bitkey system – one on your phone and one on your Bitkey device. The third is stored on Bitkey’s server and cannot move your bitcoin on its own.
- NO SEED PHRASE: Set up and use Bitkey without creating or storing a seed phrase.
- 2-of-3 MULTISIG: Three keys are stored separately across your phone, Bitkey device, and Bitkey’s server. Any two keys are required to move your bitcoin.
- BUILT-IN RECOVERY: Encrypted backup and recovery tools can help you regain access if you lose your phone or Bitkey device. You can also designate a Recovery Contact.
The analysis computed 51 indicators, including EMA 9/21/50/200, RSI, MACD, ATR, Bollinger measures, candle shape, streaks, and 15-minute and one-hour candles. Twelve indicators were selected on the validation window for the Jev comparison. Jev was tested on 1,500 randomly selected test candles in two input formats: numeric indicators and plain-text descriptions. The author reports using 4,560 Jev requests and estimates their cost at about $0.19.
Inputs were blinded, according to the author: Jev was not given the symbol, dates, or prices. Requests also included control questions answerable from the supplied indicators, such as whether RSI was above 50. Jev answered those controls correctly 99.8% of the time, which supports the claim that the main result was not simply caused by an inability to read the inputs.
How Jev compared with the benchmarks
On the full 181,290-candle test set, the author reports these results. Accuracy is the share of correct up/down classifications; AUC measures how well a model ranks positive outcomes above negative ones, and 0.5 represents chance-level ranking.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Rank #2
- Unparalleled Security: Protect your assets NDA-free EAL 6+ Secure Element, offering robust defense and complete transparency
- Simple & Secure Interface: Manage your digital assets easily with a clear OLED screen for secure on-device confirmations
- Supports 1000s of Coins & Tokens: Securely handle thousands of assets, including Bitcoin, Ethereum, and more, all in one wallet
- Effortless Asset Management: Monitor and transact seamlessly with Trezor Suite, our intuitive desktop and mobile app
- Enhanced Backup Solution: Rest assured with Multi-share Backup, eliminating single points of failure for secure cold wallet recovery
| Approach | Accuracy | AUC |
|---|---|---|
| Always predict the base rate | 49.9% | 0.500 |
| Continue the last candle’s direction | 49.5% | 0.495 |
| Reverse the last candle’s direction | 50.6% | 0.506 |
| Logistic regression, 51 indicators | 51.5% | 0.522 (95% CI 0.520–0.525) |
| Gradient boosting, 51 indicators | 51.9% | 0.527 (95% CI 0.524–0.530) |
| Gradient boosting, selected 12 indicators | 51.7% | 0.526 (95% CI 0.524–0.530) |
The smaller head-to-head comparison used the same 1,500 unseen candles for Jev and the selected-indicator boosting model:
| Approach | Accuracy | AUC | Control questions |
|---|---|---|---|
| Gradient boosting, selected indicators | 51.6% | 0.522 (95% CI 0.493–0.550) | |
| Jev, numeric indicators | 48.1% | 0.486 (95% CI 0.459–0.516) | 99.8% correct |
| Jev, text descriptions | 49.3% | 0.494 (95% CI 0.463–0.524) | 99.8% and 99.7% correct, respectively |
All three AUC confidence intervals in the small comparison overlap 0.5. That means the 1,500-candle results do not establish a statistically clear AUC advantage or disadvantage for any of the three approaches. Jev’s point estimates were below 0.5, but that alone is not proof it predicts in the wrong direction.
Why the author said Jev read the indicators backwards
The “backwards” description refers to the direction of Jev’s responses compared with the relationship the author measured in the test data. For distance above EMA9, the reported rank correlation with predictions was −0.80 for boosting, +0.83 for Jev with numbers, and +0.72 for Jev with text. For the move over the last three candles, the figures were −0.66, +0.76, and +0.70. For same-color streak, they were −0.53, +0.65, and +0.87; for RSI(14), −0.76, +0.77, and +0.51. The author also reports that boosting and Jev predictions correlated at −0.71.
Rank #3
- Unparalleled Security: Protect your assets with EAL 6+ Secure Element, offering robust defense and complete transparency
- Simple & Secure Interface: Manage your digital assets easily with a clear OLED screen for secure on-device confirmations
- Supports 1000s of Coins & Tokens: Securely handle thousands of assets, including Bitcoin, Ethereum, and more, all in one wallet
- Effortless Asset Management: Monitor and transact seamlessly with Trezor Suite, our intuitive desktop and mobile app
- Enhanced Backup Solution: Multi-share Backup eliminates single points of failure for secure cold wallet recovery
A concrete comparison in the article makes the contrast easier to see: when price was far below EMA9, 52.8% of next candles closed up; when it was far above EMA9, 46.9% did. Jev’s average prediction reportedly moved in the opposite direction, from 41.6% to 54.4%.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteThe author’s explanation is that Jev followed textbook-style momentum interpretations—elevated price, RSI, and a run of green candles suggesting continuation—while this particular five-minute BTC dataset showed a slight pullback tendency after stretched moves. That is an interpretation of this setup, not evidence that technical indicators always work in reverse or that the pattern will persist.
Why the result does not amount to a trading edge
Even the selected-indicator boosting model’s most confident 10% of calls reportedly earned 0.74 basis points per trade before costs. The author compared that with an estimated 15–20 basis points for a Binance spot round trip. These are reported figures for the study’s setup, not a current fee quote or a result that transfers automatically to other accounts, markets, or strategies. The pre-cost signal was far smaller than the stated fee estimate.
Rank #4
- Dual-chip architecture for maximum protection: The next-gen, fully auditable TROPIC01 chip works alongside a certified EAL6+ Secure Element—completely NDA-free—to deliver radically transparent, industry-leading defense against physical attacks.
- Quantum-ready security: Get protection against future threats with the first-ever hardware wallet designed with quantum-ready architecture.
- See every detail with confidence: Our largest high-resolution color touchscreen makes it easy to navigate your assets, review transactions and manage your coins with clarity.
- Wireless freedom with encrypted Bluetooth control: Manage, buy, swap and stake securely using Trezor Suite on desktop or mobile. Qi2-compatible wireless charging keeps your Trezor powered up. No cables required—security meets convenience.
- Works seamlessly with Android, iOS and desktop: Connect wirelessly or via USB-C to your phone or computer. Manage your crypto anywhere with our companion Trezor Suite app.
The author also tested three ways to combine Jev with boosting: replacing boosting, trading only when Jev agreed, and blending their probabilities. Replacing the model performed worse; agreement filtering underperformed letting boosting keep its own most confident calls; and blending produced no material improvement. The fitted Jev weights were −0.09 for the numeric format and +0.16 for text. In one comparison, numeric Jev agreed with just 6 of the 300 most confident boosting calls and objected to 294.
Those results counsel against using Jev as an order gate in this particular setup. They do not prove that every combined system, different prompt, or richer set of inputs would fail.
What the test leaves unresolved
- Model version: The reported Jev test used jev-1.13.0. A newer version could produce different results.
- Market and timeframe: The test covered Binance BTCUSDT five-minute candles, not other exchanges, assets, or time horizons.
- Input type: It used price-derived indicators, not news, filings, order-book context, or prediction-market information.
- Sample uncertainty: Jev’s 1,500-candle AUC intervals overlap 0.5, as does the boosting comparison’s interval. A larger or different test could change the estimates.
- Prompting: The author notes that specifying a reversal rule could change Jev’s response. Following a supplied rule would not, by itself, show that Jev independently discovered it.
- Reproducibility: The follow-up points readers to public code as a way to rerun the experiment, but the reported test has not been independently reproduced in the material cited here.
The follow-up characterizes the conclusion as a late-September 2026 snapshot. It could change if a newer Jev version passes the same kind of test, if a baseline comparison on unseen data after costs is published, or if Jev is evaluated on a task that includes information beyond price.
Best Value
- Effortlessly build your crypto portfolio via the all in one Ledger Wallet app: buy, sell, send, receive, swap, stake and more across popular blockchains. 15,000+ coins & tokens in a single dashboard. Keep a close eye on the market. Compare service providers. Track performance. Get timely alerts. Build your portfolio with confidence.
- Effortlessly build your crypto portfolio via the all in one Ledger Wallet app: buy, sell, send, receive, swap, stake and more across popular blockchains. 15,000+ coins & tokens in a single dashboard. Keep a close eye on the market. Compare service providers. Track performance. Get timely alerts. Build your portfolio with confidence.
- Enjoy Bluetooth connectivity, iOS access, and hours of battery use with this mobile-first, secure backup signer. Freedom you can depend on.
- Genuine Check: confirm your signer is authentic during setup with the Ledger Wallet app.
- Protect your signer: keep it in mint condition at all times with a bespoke Pod or Case to avoid scratches and everyday wear and tear.
How to assess an AI trading claim
Before treating a model’s forecast as actionable, ask questions that distinguish a plausible-sounding answer from measured performance:
- What is the baseline? Compare the model with simple rules and conventional models on the same task and data.
- Was the evaluation genuinely out of sample? Choices made after inspecting the test set can make results look stronger than they are.
- Are costs included? Spread, fees, and slippage can erase a small pre-cost signal.
- Did the model read the input correctly? Controls can test whether it extracted the supplied facts, though passing them does not prove forecasting skill.
- What happens when it is uncertain or wrong? A real deployment needs explicit limits and failure behavior, not just a confidence score.
Structured tasks such as labeling, classification, or checking a trading journal against stated rules may be easier to evaluate than predicting a future price movement. The author offers these as potentially better-suited uses for Jev, not as a general proof about all AI systems.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
