The experiment / decision evidence
Bot Analytics
Are shorts working? What changed its mind? Is it getting better?
Follow the recorded evidence—not a story written after the trade.
Loading recorded paper results…
Before the news / on the record
What do Grok and Astra expect next?
Both Super Bots use a shared economic database, with the evidence saved at each review. Each makes its own forecast and decides how much paper-trading risk it wants to take. We score the forecast separately from the bet. When a bot declines to forecast a release, we show that decision and its reason too. Every release it was asked about is listed, including any it left unanswered.
Reading the forecast records…
How are the forecasts scored?
The economic number: We compare the bot’s prediction with the first official value our system observes after release. Smaller errors are better. We also compare it with simply repeating the previous reading.
The range: The bot gives a range it expects to contain the result 80% of the time. We check both missed results and overly wide ranges. Range score = the width of the bot’s 80% range plus ten times any miss outside it. A very wide range that happens to contain the result still scores badly, so “inside its range” alone is not accuracy.
The timing: A forecast counts as early when it arrives at least an hour before the release and short-notice when it arrives 10–60 minutes before. The two are scored separately. Anything received later is kept on the record but never scored.
The market: We score the predicted chance of SPY falling in a fixed 30-minute window. For news before the open, that window is 9:30–10:00 a.m. New York time. For news during trading, it starts at release. We never credit an earlier overnight move.
The risk: Desired exposure is the bot’s plan—not a placed trade. Existing trading rules still decide what can execute. Being more aggressive does not make a forecast more accurate.
A Cleveland Fed nowcast is a separate model’s forecast, not market consensus. We do not currently have a consensus feed. Economic figures can be revised. A few correct calls are not proof of skill or profitable trading.
Check the evidence and update status
How this bot trades and learns
The bot’s description will appear with its report.
Long vs short
Paper trading · completed tradesA long trade aims to profit when the price rises. A short trade aims to profit when it falls. Each “exit” below means a trade, or part of a trade, was closed.
Is this bot improving?
A better win rate can still lose money. Different market conditions, trade sizes and partial exits can change these results. Improvement is not proof that learning caused it.
What is the bot trying to do?
Its saved explanation—not private thoughtsStart with the plain-English explanation: what the note says, what its rules mean, and what we cannot confirm. The bot's claims are not independently verified facts. Open the original note for every detail.
Quick guide to trading terms
RSI: Measures the strength of recent price moves. A high reading can suggest a stock has risen quickly; it does not guarantee a fall.
EMA / SMA: Moving averages that smooth price changes to help show a trend.
ATR: Measures how much a price has been moving. It helps describe volatility.
Resistance: A price area where previous rises have struggled. It can still be broken.
Trailing stop: A stop that follows a favourable price move. It does not guarantee the exit price.
Shadow mode: A rule is being tested and recorded, but its suggestion may not control the trade.
What the bot says it has learned
Saving a lesson does not prove the bot used it—or that it improved results.
Did the bot change its rules?
The bot can review new information and still keep its existing rules. “Short / medium / long” names a time-horizon rulebook—not whether it buys or sells short. A five-minute price bar covers five minutes of market activity; it does not set the length of a trade.
What information did it record?
These counts show what was saved with decisions. They do not prove the information caused a trade. Two records can refer to the same trade.
We do not have a complete record linking company valuations and individual headlines to each order. For example, we cannot say a bot shorted an oil stock because of a peace deal unless its records show that link. That is a question to investigate, not a result.
Decisions during the day
Did the bot's adjustment rules help?
Some rules suggest larger trades, smaller trades or different stops. This table compares those suggestions with later results from the past seven days. Test-only suggestions and active rules are included together, so it cannot prove a rule helped. Returns include estimated costs only when those costs were recorded.
How to read this: A positive percentage means that group of recorded trades had a positive average return after estimated costs. It does not show how much extra profit the adjustment caused. “No size change” means the rule did not suggest changing the trade size—not that the bot did nothing.
Recently closed trades
Showing up to 40 closed trade records. Closing part of a trade counts as a separate record. The totals above use all available records in your chosen period, not just these 40. A profit counts as a win; breaking even does not.
What made it close? The last column explains the saved exit reason. That reason alone does not tell you whether the trade won or lost—check its paper profit or loss. A dash means a usable figure was not recorded, not zero.
Simulated money, real market prices. Costs are estimates where recorded, not brokerage charges. Bots start on different dates and use different risk limits. No claim of a complete market picture, profitable learning or future returns. Not financial advice.