Anais Dotis-Georgiou explains that classical machine learning methods such as Error, Trend, Seasonality Forecast (ETS), Autoregressive Integrated Moving Average (ARIMA) and Holt-Winters are powerful time-series predictors that outperform many other ML methods including Long Short Term Memory (LSTM) and Recurrent Neural Networks (RNN) in One-Step Forecasts. The author focuses on Holt-Winters for three reasons: it's a sibling of ETS, can be used out of the box with InfluxDB, and the InfluxData community has requested an explanation of Holt-Winters. The article is divided into three parts, covering when to use Holt-Winters, how Single Exponential Smoothing works, and how to optimize its parameters using optimization techniques such as minimizing the Residual Sum of Squared Errors (RSS).