I would imagine that it has more to do with the existing ecosystem, the swap between trident and the custom USB-C controller they use in the other devices and m

I would imagine that it has more to do with the existing ecosystem, the swap between trident and the custom USB-C controller they use in the other devices and mechanical changes in the form factor internal to the phone. If you nuke lightning from iPhones, you have the same problem as with the 30-pin connectors: entire cars suddenly don't work with your phone anymore, many years worth of produced lightning devices no longer connect and everything from charging and data exchange to debugging and alternate output modes will no longer work. This is not a technical problem to solve, since they already wrote tools and created workflows for doing that on the less risky product lines (Macs, iPads etc.) where the impact of a bad port implementation is much lower. But phones are still carried around all day (in pockets no less), charged/plugged a lot etc. which while seemingly slightly different makes for additional risk when implementing. Considering that seemingly insignificant changes (like that dumb antenna shorting problem when holding the phone in the iPhone 4 due to a small case change) can lead to huge and very expensive problems (like replacing all panels on the first white iPhones due to the paint not sticking well enough). If you have 70 million iPads and 1% has the issue, that's not great but also not too bad. If you have 200+ million iPhones and 1% has an issue, that's pretty bad, and you also get a lot more flak because people are very emotional about their phones, as well as phones they don't have and would never buy but still complain about. Back to the supply chain and logistics, all fixtures, machining workflows, testing and component sourcing would have to change, which isn't too bad on its own, but because there are still many many years of supported phones out there, you have to keep the old method around as well. This problem doesn't exactly go away, regardless of how long you stall, but if you work on a method to make that less of a problem (i.e. make sure the next revision of test harnesses and repair fixtures account for it when they get cycled), while also growing your ecosystem (other devices with USB-C within your brand) you can build up knowledge and specialism in your workforce (i.e. the people that make the manuals for the people that work in the store can lean from inevitable mistakes they will make in the initial USB-C versions of iPads). That means that when you do finally make the seemingly tiny change (which for charging really is just a bunch of B.S. since it's not about the port on the device but the plug you stick in the charger) and get those internal USB-C models they probably have had for years up in the design for manufacture and retail phase, the risk is much lower. It is however, just as likely that they will go full iPhone X again and make a portless phone instead. Going portless has the same ecosystem risk (mostly marketing risk) since they would be breaking compatibility with lighting anyway, but because there won't be a new port, there isn't going to be a change in material thickness, ingress points in general or new chips to add to the design. So perhaps that would be half the risk of another connector migration.