Statistical Stock Price Forecasts for Swing and Position Traders
TSFStocks forecasts specific limit order prices for each trading day, published one week in advance. Historical validation shows an average of 80% probability of profitability within 120 days. All forecasts and results are available for independent review.
TSFStocks is a research platform. Nothing on this site constitutes investment advice.
TSF forecasts generate daily limit order prices for each stock, a week in advance.
A trade event occurs only when the market's intraday low actually reaches the forecast entry price. Wins are capped at the 5% profit target — once the intraday high hits the target price, the position exits. A loss occurs only when the position fails to reach the profit target within the maximum hold period.
The Top 10 Swing Stock Picks use a 5% profit target with a 7-day maximum duration. The Top 10 Position Stock Picks use a 5% profit target with a 90-day maximum duration.
Click any ticker name to see its full validation history from 2021 through today.
Top 10 Swing Stock Picks
Top 10 Position Stock Picks
Not a Backtest
TSF forecasts are built from the historical close prices of each stock, and the daily limit order prices are generated one week in advance. All trade signals used for validation use the intraday low and high prices — an entirely separate data stream. The forecast model has never seen the prices it’s trading on.
TSFStocks is a research platform. Nothing on this site constitutes investment advice.
Forecast Methodology: TSF vs Legacy Models
Legacy forecasting tools, used in every forecasting platform and including everything from simple moving averages to the most advanced AI-powered machine learning models, are univariate (they predict a single value using only the historical values of that variable) and one-dimensional: they can only tell you WHAT the next number will be. TSF is univariate and two-dimensional: it tells you both WHAT the next number will be and WHEN it will change. Legacy forecasting tools can't tell you WHEN to act because legacy forecasting tools are one-dimensional and can't consider the dimension of time.
The TSF forecast methodology is Temporal Structural Forecasting. This approach is based on the Model of Temporal Inertia, which forecasts along two timelines rather than one. Complex irregular seasonal models act as lenses in a microscope for time, revealing time-based structural patterns in data that are impossible to see with the naked eye. Historical patterns of accuracy of each forecast strategy create a "map of normal" that describes what the actual values should be each day.
When actual values fall outside of the green zone confidence bands, it's not an error, it's a signal. Conditions right now are not normal. When intraday prices on a stock fall below the confidence band value, it's a signal that the price of the stock is abnormally low. Extensive research has shown that a limit order filled at this price has an 80% chance of being profitable within 120 days.
The forecast methodology and the configurations of entry and exit signals used to generate the results presented on this site are presented in the Methodology section. Preprint results of the preregistered research studies on these data are available in the Research section. All historical data on this site may be downloaded by registered users (with a free account) for independent validation and testing.
Signal Subscriptions
Forecasted limit order prices are available by subscription for up to 50 tickers. Updates are published every Sunday with a daily limit order price for each ticker based on the selected forecast strategy and confidence band. Subscriptions are for individual, non-commercial research use only. Plans and pricing →