There's also a newer book by Hastie (and Efron!), which I very much prefer to The Elements of Statistical Learning: Computer Age Statistical Inference.

https://web.stanford.edu/~hastie/CASI_files/PDF/casi.pdf

It's really well motivated and, unlike ESL, discusses many different schools---including classical inference, empirical and Bayes deep learning. Without these different perspectives, newcomers often find statistics very obscure as it just looks like a bag of tricks.

And to close the triad of machine learning bibles, don't forget Murphy's, which will apparently be extended soon! https://probml.github.io/pml-book/

I also like Alpaydin's Intro to ML even though it's not as famous: https://mitpress.mit.edu/books/introduction-machine-learning...

Another hidden gem is Webb‘s Statistical Pattern Recognition https://www.wiley.com/en-us/Statistical+Pattern+Recognition%...

Why do you think it's a hidden gem?

So much of that book just goes over my head while I didn't have that problem with ESL. I don't know if it's Murphy's writing style or just the way he approaches the topic but I found his book significantly more difficult to process.

On the contrary, I like Murphy and cannot stand ESL. It probably boils down to what statistical camp you are more comfortable in.

My problem is more with understanding, not necessarily with liking or disliking either book.

Yes! They are just like the Bible. Concise, applicable, on the point, and you can learn so many useful (and true!) things from them.

I love this book so much. It takes a strong Bayesian point of view that makes things so clear to me. It's well written and we'll structured. It starts with a summary chapter of ML which honestly by itself gets you to a very good place in understanding the basics of ML.

I remember putting this book in my pile almost a decade ago, is it still relevant?