I think that this is where you're going wrong. Spare turns aren't intended to provide "protection" on the day - they're there to be used as part of the rostering process to cover for known absences (at the rostering stage) such as annual leave, sickness, training, investigations etc.. The number of spare turns should, through an agreed ratio, be set at a level that just about covers all that on a typical day, so all the spares should be allocated to jobs and no jobs left uncovered. There's no such thing as a typical day, though, so you'll inevitably find that there are bad days where they've got no spares left at all and jobs still uncovered, and good days where they've still got spares left over. I don't think that a day where you've got spares left unused indicates that the spare ratio is wrong - it's just a symptom of the fluctuating nature of absence. It's undoubtedly exacerbated by route knowledge deficiencies (caused by the Covid training backlog or otherwise) and reduced service levels at some depots.
The bit about "protection" sounds more like it should relate to service recovery, standby, cover turns - the name varies. They're usually a booked job in their own right (and thus on the other side of the spare ratio calculation). They're there to rescue the job if it goes pear-shaped on the day. If there are any spares in addition to that, it's a bonus, but you can't rely on it. That's why it isn't really a problem, in my view, for the spares to remain at home. The service recovery traincrew are there if they need someone quickly, and in my experience that's swiftly followed by a telephone call: "I've had to use the standby - can you come in and sit spare please?". If someone in the messroom does have Covid, then at least you're minimising the proportion of the depot that's likely to be wiped out as a consequence.