> Replacing 225 mph trains with 400 mph maglevs should be worth it, especially on long routes like the 800 mile trip between Shanghai and Beijing.
That's 2h vs 3.5h, not taking into account acceleration. Time on a train can be used productively, so it isn't lost. Taking opportunity costs into account, it may or may not be worth it. I'm also not sure if 225 mph is the physical limit of conventional trains.
Also, faster trains don't necessarily make the network faster. Especially when using an integrated time table, you can run trains slower (or let them wait, but that has the same effect and wastes energy) to improve connections, thus reducing waiting times and ultimately getting people faster to their destinations. I don't know how China's railways work in detail though and they may very well optimize for making Shanghai-Beijing fast.
> China, however, has no problem doing large infrastructure projects.
That may change when they have to spend more on maintenance as their infrastructure grows and ages. Also, the question is not whether they can replace their HSR with maglevs -- of course they can -- but whether the party considers it the best use of resources, also compared to other large infrastructure projects.