The suggestion appears to be that the level of service be reduced until all trains are crush loaded, at which point everyone will abandon their cars in joy.
As an exercise, I've taken the ScotRail line-by-line financial figures posted upthread. and done some estimating.
If the average seat-mile utilisation in each of the four service groups (Greater Glasgow, East and Central Scotland, Intercity, and Rural and NE Scotland) were to increase to the maximum in that service class, and that revenue was to increase proportionately, the net subsidy required would go from £244 million to £60 million. More than half of that would come from increasing ridership on the Edinburgh-Glasgow route, the North Electrics, and the Argyle Line. There'd also be seat loadings over 100% - i.e. routine standing - on the latter two routes and the Cumbernauld-Falkirk line.
What
@NCT seems to be suggesting is (e.g.) reducing the E&G trains from eight cars to something like three cars (bringing seat utilisation up to about 100%, with lots of standing at peak times). Doing the numbers - I'm assuming 100% seat utilisation for 'short' journeys and 80% for 'long' ones, more or less judged by eye but generally the former is suburban while the latter is intercity and rural, and no savings other than rolling stock - that brings the subsidy down to about £125 million.
In other words, it looks broadly like there's more benefit to the railway in figuring out how to fill the seats it has, than in reducing the number of seats to suit the number of passengers.
By comparison, closing the Far North, West Highland and Kilmarnock-Stranraer entirely would reduce the subsidy from £244 million to, erm, £225 million, even assuming no loss of contributory traffic. Hardly sums worth shouting about.