AI trading forecast tool: what to look for, and where ApolloAlpha fits

Many AI trading products are sold on promises. If you are comparing tools, or looking for an alternative to the one you use now, the better question is: what evidence does it show, and can you check it? This page is a checklist, followed by how ApolloAlpha answers each point.

1. Does it say what it predicts?

A clear tool states one question and answers it the same way every time. ApolloAlpha answers one: will last week's high or low hold or be breached this week? Each market gets a probability, not a vague "bullish" label.

2. Is the track record tested on data the model had not seen?

A model that is scored on the data it learned from always looks good. Ask for out-of-sample results. ApolloAlpha's record covers 164,485 weekly calls from January 2016, tested walk-forward with no look-ahead, then independently re-implemented and audited. 72.7% were right, and 82.6% when the model was at least 70% confident. This is a historical test, not live trading. See the method and what it does not show.

3. Does it show how confident it is?

An 85% call and a 55% call should not be treated alike. Every ApolloAlpha forecast comes with its stated probability, and the model is checked for calibration: when it says 70%, it should be right about 70% of the time.

4. Does it separate accuracy from profit?

Calling a level correctly is not the same as making money. Costs, timing and position size all matter. ApolloAlpha says so plainly: it is not a broker, adviser or signal-selling service, and it never tells anyone to buy or sell.

5. Can you try it before paying?

The free plan shows last week's forecasts for 7 tickers, so you can compare them with what the market actually did. Paid plans add live forecasts and more of the 160+ markets. See the markets covered and the plans.

Frequently asked questions

What is an AI trading forecast tool?

Software that uses a statistical or machine-learning model to estimate what a market is likely to do, usually as a probability. ApolloAlpha estimates the chance that last week's high or low holds or breaks, for 160+ markets, every week.

Is ApolloAlpha a trading bot or a signal service?

No. ApolloAlpha does not place trades and does not tell anyone to buy or sell. It publishes probabilities only, which people can use alongside their own analysis. It is statistical model output, not financial advice.

How accurate is it?

Across 164,485 weekly calls tested out-of-sample from January 2016, 72.7% were right, and 82.6% were right when the model was at least 70% confident. This is a historical test, not live trading, accuracy is not profit, and past results do not guarantee future results.

Which markets does it cover?

More than 160 stocks, ETFs, leveraged funds, commodities and crypto. The full list is on the markets page.

Is there a free plan?

Yes. The free plan shows last week's forecasts for 7 tickers. Paid plans add live forecasts and more markets.

This is statistical model output, not financial advice.

← Back to ApolloAlpha