Patterns and non-Gaussian textures are ubiquitous in astronomical data but challenging to quantify. I will present a new powerful statistic, called the “scattering transform”. It borrows ideas from convolutional neural nets (CNNs) while retaining the advantages of traditional statistics. As an example, I will demonstrate its application to weak lensing cosmology, where it outperforms classic statistics and is on a par with CNNs. I will also show interesting interpretations of the scattering statistics. I argue that the scattering transform provides a powerful new approach in astrophysics and beyond.