A “Dashingdon style” save system is actually one in which players can save their progress at any point (not just at author-selected check points) and later load that save at will. Unlike chapter checkpoints, it requires Javascript edits, and can’t be included in published games. Including it would require a rethink of the line on back buttons, since it’s pretty much the same thing.
I doubt that refund-requesters who speed-read the whole thing within the hour outnumber refund-requesters who make a choice they dislike, realize they’d have to re-read the whole game to alter it, and rage-quit. And I’d be surprised if either group is substantial enough that CoG should be making major business decisions with them in mind.
The real business question this raises is I think not about refund rates but whether the core experience CoG sells would be compromised by adding a save-at-will function. The answer isn’t obvious; we can see from this thread that CoG readers disagree on what the core experience is and/or whether a save point would affect it.
Like I said last time we talked about this,
As a way to get more data on customer preferences, I suggested that CoG might want to give Hosted Games authors the option to add save-at-will to their games. HGs aren’t bound to deliver anything resembling a “core experience”. Giving authors the option to add save slots would give us a point of comparison, both within HGs and between HGs and CoGs, on this key business model question.
(Also, on that thread I voiced my support for the kind of checkpoint-facilitating changes that @Doriana-Gray is asking for above.)