In terms of the general slow down of bus services, TfL are now reporting that bus use is starting to fall again - the fact that for short journeys at peak times it's often quicker to walk is assumed to be one of the factors.
(From data that's public domain, route 11 for example now has an average end to end speed of slightly less than 4 1/2 mph in peak hours)
The number of new timetables over recent months to try and improve reliability has been huge (more info
here) - the curtailment of the 53 at Lambeth North is an extreme, but on many routes, extra buses have been added, or the headway slightly reduced (or both) to allow for longer (and longer) journey times.
Doing this is not a simple process - in London, TfL have to agree it (and where relevant, the additional costs) and the operator has to get a driver schedule that works. This is not always as simple as you would think, as that driver schedule has to fit within drivers' hours rules (these are on the basis that the driver MUST have a break after X amount of time, not that the driver is entitled to that break.)
As a hypothetical example, it may have been that it was possible to do 2 round trips on the 53 before the driver was legally required to take a break - now it's not quite possible if the route continues to run its full length.
Drivers are still employed on (say) an 8 hour day. If they can only now manage 2 trips on the 53 in the course of a day, you either get duties where drivers are paid for not doing anything, or you have to make arrangements for breaks somewhere else (and that means unproductive time for at least some drivers to travel there to start sequences of breaks - you can't just have drivers abandoning buses at the roadside, or leaving them at remote termini for breaks), or you drag more routes in to the process so that drivers work part of their day on the 53 and part on something else. That in turn has potential route / type training issues.
In this case, it's probable that curtailing the route was the easiest answer. (this is all speculation on my part - I am not involved in any way with route 53)
In terms of managing 'short turns' once buses are on the road - yes, you could argue that they should all be avoided, but what if there is a bunch of buses one way and a gap the other? what if one of the drivers who is running late is going to run out of hours and not be able to do their return journey?
How it's done can certainly be improved - I had an experience a few weeks back where the bus I was on got to within sight of the bus in front (which was probably running even later than the bus I was on), then my bus was held back 'to regulate the service' (the bus in front slowly disappeared in to the distance) then it was turned short. Hmph.