> Looking at graphs with clear improvements on many metrics .. one of the UK's most hated things the government has ever done

This is a classic of metrics driven management, isn't it. Customer satisfaction isn't in the metrics so people stand around saying "well they should be happy!"

Mainly about the cost, but the overcrowding on certain lines and the unreliability are also major factors. Commuters end up using the train because they're more or less forced to - you're not going to drive into rush hour London, and there's nowhere to park; and that's also a matter of policy to get people on the trains for sustainability reasons.

The "playing at shops" market setup exacerbates the problems. Original comment says "costs have gone down due to competition", but since the UK's operators are little regional monopolies they don't actually compete. The operators own neither track nor rolling stock; most of the money flows to the "train landlords", the ROSCOs.

Privatising the track, through "Railtrack" was the big disaster. They cut maintenance, but when you enshittify railways eventually people get killed, and nationalization was inevitable after Hatfield.

They managed to increase rail's share of overall transport. Reversing long running trends from before.

> Commuters end up using the train because they're more or less forced to [...]

Rail transport had been losing passenger share. Did the private companies somehow figure out how to 'more or less force' people to take the train? Why couldn't the government rail do that before?

> Why couldn't the government rail do that before?

They didn't want to. Privatisation was the era of Thatcher and Major, the discovery of North Sea oil reserves, and of the same general desire to culturally align with US (including car culture) that to the uban design found in Milton Keynes 30 years earlier.