Will support hold?
Will resistance break?
Know the odds.

Yes, there's a free plan (last week's forecasts for 7 tickers), and you only need an account. The Live Trial adds the live forecasts for $2.49 a week at the founders price ($4.99 a week list). Founders pricing also applies to the paid plans until January 2027.

Sign up before January 2027 and you pay the founders price (for example Standard at $24.99 a month instead of $49.99) for as long as you stay subscribed. On the monthly plans it also gets 5% cheaper every month you stay, down to 60% of your starting price. Anyone who signs up after founders pricing ends pays the list price.

The founders price is per sign-up. If you cancel and later resubscribe, the monthly 5% drops start again from the starting price. Founders (accounts from before January 2027) start again at the founders price; everyone else pays the list price (Standard $49.99, Pro $64.99, Institutional $79.99 a month, or $4.99 a week for the Live Trial).

Yes. Create a free account with Google or email and you can use the dashboard straight away.

ApolloAlpha is designed as a quantitative market analytics platform that publishes statistical estimates of predefined weekly price events, with backtested track records. It does not personalise forecasts to your financial circumstances, investment objectives or risk tolerance, and it does not issue instructions to buy, sell, hold, enter or exit investments. The forecasts are intended to support your own independent research and analysis. Past accuracy does not guarantee future results. Whether a particular service falls within financial-advisory regulation depends on its features, presentation and applicable law.

160+ markets spanning large-cap stocks across every major sector, ETFs, leveraged funds, commodities, and crypto. The full list is on the Markets page, and the Symbols by plan page shows which symbols each plan includes. Don't see the symbol you're interested in? Institutional members can request additional symbols. Requests are considered for our research coverage: where the data allows, we may add the symbol, develop a model for it and share its historical test results. Not every request can be added.

No. Each forecast comes with a plain-language description of the estimated outcome, its probability, and the track record behind it, whether you're new to markets or you've been trading for years.

No. ApolloAlpha publishes statistical estimates of predefined weekly price events, with the out-of-sample track record next to every forecast. It does not issue trade instructions or tell you whether to buy, sell, hold, enter or exit a position, and it does not execute trades or connect to brokerage accounts. It is a proprietary statistical model developed in-house, the same model our firm uses for its own forecasting and trading. You are responsible for evaluating the information and for your own decisions.

ApolloAlpha is built by a team of engineers and quantitative analysts who graduated from Ivy League universities and have institutional trading and quantitative research experience.

Support is a price level where a market has recently stopped falling; resistance is a level where it has recently stopped rising. ApolloAlpha uses last week's low as support and last week's high as resistance, and forecasts the probability that each one holds or breaks over the coming week.

Weekly. Each forecast is produced after the Friday close for the week ahead, and the previous week's outcome is added to the track record.

No. The free plan needs only an account (Google or email). You add a card only if you choose a paid plan.

It varies by ticker — that's the point. Instead of one headline number, every ticker shows its own accuracy, AUC, Brier score, and full confusion matrix, so you can judge confidence yourself rather than take a marketing claim on faith.

It is the model's estimate that the level will hold, for example 80% that last week's low holds as support. It is a probability, not a promise: at 80%, roughly 8 in 10 similar calls should hold and 2 in 10 should break. Calibration checks whether that is true in practice. It is the probability of that defined price event, not the probability of making money: costs, execution prices and gaps affect trading results, and any single forecast can be wrong.

They are two views of the same thing. The ROC curve is the picture: as the confidence cut-off changes, it plots the breaches the model catches (up) against the false alarms it raises (across). The AUC is the score read off that picture, the area under the curve. Put another way, take one week that held and one that broke: AUC is the chance the model gave the held week the higher probability. 0.5 is a coin flip (the diagonal line) and 1.0 is perfect. A curve that bows toward the top-left corner has a high AUC.

The Brier score measures how close stated probabilities were to what actually happened. 0 is perfect, and always answering 50% scores 0.25, so lower is better and a score well under 0.25 means the probabilities carry real information. It punishes being confidently wrong, which accuracy alone does not.

A 2 by 2 table of what the model called against what happened: calls that held and did hold, calls that broke and did break, misses, and false alarms. Large numbers on the correct diagonal are good. We show it for every ticker so you can see where the model is strong and where it is not.

A forecast is calibrated when its stated probabilities match reality: of all the times it said 70%, about 70% happened. The reliability curve in the dashboard plots exactly this. Points near the diagonal mean the stated odds can be taken at face value.

Out-of-sample means tested on weeks the model had never seen when it made each forecast. Every forecast uses only information available before that week, and our headline track record counts weeks from January 2016 onward. This is what separates a real track record from a backtest that peeked.

See how out-of-sample testing works, step by step →

Two ways to read the confusion matrix. Precision asks: of the weeks the model called "hold", how many really held? Recall asks: of all the weeks that really held, how many did the model call? High precision means few false alarms; high recall means few misses. The same two measures apply to breaches. Looking at both stops one flattering number from hiding a weakness.

The correlation table compares how closely two markets' weekly returns have moved together: +1 moves in step, 0 is unrelated, and -1 moves opposite. It uses ordinary (Pearson) correlation of weekly returns over a chosen window. Short windows are noisy, so the 1-year and longer views are steadier. It is a diversification aid: calls on markets that move together are less independent than they look.