I appreciate your work here and I am sorry to make you spend your valuable time on such an inconsequential issue. If I had to bet I would say you are probably right.
Still, I am not totally convinced. I find it odd that:
1) Another player with 2k salary (500 more) and the same game shape who wasn't on the DC at all played 12 minutes.
2) I used to use multiple sub 1k salary players and they were always put in the game with a large 4th quarter lead (even using "coach picks").
It would ease my insatiably curious mind if you could put these points in context. Again sorry for wasting your time and this will be my last post on the subject.
Although I'll of course defer to Marin's analysis if he should say otherwise, I think the clear result here is:
While the game was not in garbage time, the coach picked the best player on the depth chart at the position, so the scrub player listed as the "starter" was never selected.
When the game reached garbage time, that overrode it so that the coach would try to ensure that no starters played. Because the player was listed on the DC as a starter, even though he never played, he was considered a priority to replace and since you had a player listed as a backup to that position, he was therefore confined to the bench.
The thing to keep in mind here is that garbage time is not a call for coaches to choose their least-effective lineups; the goal is to relocate the starters to the bench. I personally don't know if there's a preference to backups or reserves, just that whenever possible a starter is nailed to the bench. The scenario that occurred here is unfortunate, and does demonstrate that substitution options could definitely be improved, but given the way LCD and garbage time works, this is exactly how this scenario should be expected to work.