Assuming that there's very little difference to the NR costs regardless of the number of trains which run along a line, the busier the line the more the NR costs are spread amongst those using it the easier it is to recover the costs.
For example a line with 1tph still has to be inspected just as much as one with 2tph (even if it's more regularly it's unlikely to be twice as registry), or to have a signaler controlling the track (other than first last may be earlier later) 6tph is no more cost than 1tph.
It is course depends on who the risk is applied.
DfT agree a minimum service level it costs them 20 for the TOC to run that service.
DfT agrees a +A service level it costs them 20 plus any losses for that +A costs but doesn't recover.
DfT agree a minimum service level but with TOC permitted to take risks (i.e. TOC is the one who loses money if they run lose making services) with a profit sharing mechanism.
The government would be scared of the second, but would have no issue with the third. As the third means they pay out a maximum of 20 but could get something extra back if the extra services do well.
I'd be interested to see the data for the -30% on income pretty passenger per km, I wonder if it is only looking at the latest quarter or if it's based on the last 12 months of data. As the latter could be much worse than the former, as Q3 of 2021/22 had Covid restrictions (at least Christmas parties being cancelled), so would have been hard to attract passengers.
If you look at the rail use data for November WFH allowed a lot of rail travel to be paused for the strike days, but then saw spikes afterwards. Almost as if people were going to the office when the strikes weren't on.
www.gov.uk
That graph also implies an upwards tend between Q2 (July to September) and Q3 (October to December), even though the strikes are ongoing.
If the strikes were to be resolved it's likely that rail use would see a reasonable increase in use (and quite possibly over what would be expected if you just tallied up the bin strike days usage).