I dislike the language of "scam", but get the point about both regular interval services and the impact of recovery time. Having lost a 32 minute late DR to 5 minutes of recovery time not used at King Edward Bridge, I get the frustration - but think the real issue is not the interval but the operation of recovery time. I would also welcome a rule of thumb that, if you've caught the +30 or +60 train on a regular interval service, you would qualify.
I moved to SE London in 2011 and over the period between then and the Elizabeth Line opening, I documented the games SouthEastern played to try to goose the performance statistics. The timetabled journey into Cannon Street kept getting longer, simply because they were pretending the trains were due into Cannon Street later than they really were. We're not even talking about a few minutes of recovery time, it got to the point where necessary turnaround time was being used to pad the timetable, so the only way a train was going to leave CST on time is if it arrived early and therefore had sufficient time to be turned around.
This practice of having one timetable for the signallers and drivers to adhere to and a different one for journey planning, performance stats and delay repay also contributes to people thinking, not unreasonably in my view, that the TOCs aren't treating them fairly and reasonably when they are delayed. Regular travellers know when their train is due to arrive: it's due to arrive when it normally arrives. It's not due to arrive at some later time specified in the timetable: that's just a fiction to avoid the first few delay minutes being recorded.
It's bad on a number of different levels: journey planners can't assume trains will arrive when they usually arrive but have to assume they will arrive at the published time, so they can't give passengers connections that are perfectly sensible. This makes the railway a slower option than it needs to be.
It's bad because, while perfect performance isn't likely or possible, measuring all the delays is important if you want to analyse all the causes of all the delays. Making the yardstick shorter in order to hide shortcomings mean the causes of small delays aren't identified or learnt from.
And it means when things go wrong and a regular traveller who normally gets to work at 9am finds themselves arriving at 9:20am, the rail industry ends up explaining to that passenger that they imagined the first 7 minutes of that delay and in fact they were only 13 minutes late.
I, too, would accept raising the thresholds if the gamesmanship would stop. Stop padding the arrival times at the terminus. Stop pretending delays didn't happen because the passenger made an unofficial connection before the delay happened. Reduce the padding on the MCTs. Use some common sense if a passenger is claiming for a sub-MCT connection but the tapping data on their season ticket shows it is a connection they always make when trains are on time. Basically, stop rejecting perfectly reasonable claims for genuine delays based on technicalities.
And stop ignoring what claimants say and overpaying for delays you think should have been longer than they were because an MCT was fouled, but the passenger has told you they made the connection.
Once all that is done, then I'd be prepared to concede that standardising at a higher number (maybe DR25) would be reasonable.