It wasn't the copying, it was the culture of AB split testing everything that killed them. AB (or multi-variate) testing can help improve an already successful

It wasn't the copying, it was the culture of AB split testing everything that killed them. AB (or multi-variate) testing can help improve an already successful product, but I'm not convinced it can build a good game from scratch (or from copy). A culture of design by AB split testing killed any chance they had at creating another breakthrough game. Sending out emails to select users about possible future games, and counting click throughs as votes, with the idea of building the game with the most votes is probably doomed to failure. They double-downed on that strategy, brought in a bunch of MBA types (not evil people, but not great game designers either) to statistically manage game development. It can't be too surprising that great games aren't built that way, and a game company without great games can't survive forever.