My personal theory is that just it's not as hard to build an electric car now as it was in 2012. Remember that back when the Model S was newly released, Tesla

My personal theory is that just it's not as hard to build an electric car now as it was in 2012. Remember that back when the Model S was newly released, Tesla seemed like they'd done something impossible. Not only had they released a practical electric car, but it was good. They'd solved a ton of hard technical problems that other companies hadn't been able to: Fast charging, liquid battery cooling, ultrasonically welded fuses on every cell, OTA software updates, permanent magnet synchronous reluctance motors, you name it. It looked like pure magic to investors. As soon as it became clear that the company would survive, the stock price went vertical. Over time, though, this niche problem space has become well understood. Legacy automakers have shown that not only can they solve the same set of problems, but they can also apply a hundred years of manufacturing experience, support, and brand loyalty. Electric cars also tend to be even simpler to build, with a smaller BOM and fewer precision-made components. In the meantime, it seems like Tesla hasn't been able to shore up their quality control issues. Tesla's remaining moat is in their software, as well as a series of longshot bets that (to me, anyway) seem less likely to pay off: Full self-driving, in-house cell manufacturing, rooftop solar, humanoid robots. At one point, I would have figured Tesla's secret sauce applied to just about any problem. Now, I'm not so sure. In retrospect, it looks like it just applied to a series of discrete engineering and industrial design problems. What once looked like Apple levels of product leadership maybe now looks more like Blackberry: An early entrant to a huge market, but ultimately not able to fend off the legacy players.