But is it worth the price? I concluded I'd rather build a vehicle that's simpler and better than the one I could buy. And it's costing me about 20% as much to do so. And it'll burn a hell of a lot less fuel.
Simpler? Sure. Better? I spent years in garages modifying and building street legal vehicles. On very few metrics were they better than what you could buy, then or now. If better means quicker, sure, you could do that. But it won't be safer, cheaper, more reliable, more efficient, lower polluting, or more comfortable than what you could buy from Ford or Toyota or Subaru or Tesla.
Modified vehicles can be completely reliable if you do it sensibly. More reliable, even, than factory. While the IDI 1HZ isn't the most optimal platform, I'm building my engine with 1HD-T rods aftermarket upgraded pistons which have proven successful in many other 1HZ-T builds. At the conservative no smoke 20psi tune I plan to run I'm confident it'll be fine, and I'll run a temp sensor in the head, EGT, and A/F meters to be sure. I'm using an efficient turbo (HX-30) with a massive FMIC. I'm building the A442F with upgraded valve body and torque converter, as well as a proven standalone controller. Running a large A/A transmission cooler as well. I'm not worried about reliability. I've also done a bunch of this stuff in other vehicles in the past and learned what works and what doesn't.
One thing I'm super excited to play with on this build is an exhaust brake. The transmission controller I plan to use will accept shift and TC lockup requests via CAN, so I'm planning on using a microcontroller to act as my "engine brake ECU" which will request downshifts, TC lockup, and increased line pressure when the brake is engaged. It'll also handle low speed shutoff. When the brake is disengaged it'll actuate the TV cable (using an extra cruise actuator and some power circuitry to run it) proportionately to boost. I'm not against computers in automotive projects in principle, I just don't much like ones I can't control.