Odd reasoning in a father vs. mother name-change case

As I understand it, American law had in the past often provided that a child would bear the last name of the father when the last names of the parents differed (and where, if the parents were unmarried, the father had officially acknowledged paternity). Today, given the court's reading of the equal protection clause as generally banning sex discrimination, the rule has largely changed to an ostensibly sex-neutral "best interests of the child" standard. But of course that standard is famously hard to apply, especially in a sex-neutral way.

Here's an odd Indiana Court of Appeals decision (Newcomer v. McQueary) that I just read. Amanda Newcomer and John McQueary never married, and their son was given the last name Newcomer at birth; but a year later, McQueary petitioned a court to change the boy's last name to his own, and the trial court agreed. Amanda Newcomer did not appeal, but kept using the name Newcomer on the son's records.

Then, when the boy was 5 years old, Amanda Newcomer petitioned to change the name back. The trial court rejected the petition, and the appellate court agreed. Part of the reason was simple civil procedure—Newcomer failed to appeal the original decision, and can't now get a second bite at the apple. Nor was the court swayed by the adjustment required to change how the child has been called for five years:

We are not unmindful of the fact that consistency is important to children and that it will take some time for C.M. and his community, friends, and family to acclimate to identifying him with the last name "McQueary." However, we find Newcomer's willful disregard for the trial court's earlier name-change order severely damaging to her argument. In essence, by flouting the trial court's order, Newcomer has engineered the predicament from which she wishes C.M. to be rescued. Had Newcomer simply complied with the trial court's order, C.M. and his community would not know him by the last name "Newcomer," and using the name "McQueary" would pose no challenge.

But then the court concluded with this:

We conclude that the evidence is sufficient to prove it is in C.M.'s best interests to retain the last name "McQueary." "Newcomer" is the last name of Newcomer's former husband, a man with whom C.M. has had no relationship whatsoever. Although C.M.'s surname will differ from that of his mother and siblings, we believe he is better served by having the same name as the father who loves and cares for him than he is by having the name of a man with whom he has no affiliation. The trial court did not abuse its discretion when it denied Newcomer's petition to restore C.M.'s name.

But to the boy—and to most other people who come across him and his mother and his siblings—"Newcomer" is not primarily the last name of the boy's mother's ex-husband; it is the last name of the boy's mother. It makes no sense, I think, to frame the dispute as "the name of a man with whom he has no affiliation" vs. "the father who loves and cares for him." Rather, the dispute is "the name of the mother who loves and cares for him" vs. "the father who loves and cares for him."

Nearly every last name, after all, is the name of many people, not just ex-husbands but parents, siblings, cousins and others (many of whom have no affiliation with the child). Why focus on Newcomer's husband as the person who has "the name" Newcomer, rather than on Amanda Newcomer herself, who is just as much a bearer of "the name"?

Again, I think the court's procedural reasoning would have been adequate—if there's a court decision and the losing party doesn't appeal it, the losing party generally shouldn't be able to undo it four years later. But the court's ex-husband vs. father framing at the end of its opinion, rather than the sounder mother vs. father framing, strikes me as mistaken.