Game Review Flaw?

First time posting, so forgive me if this is in the wrong place. I reviewed two of my games recently where chess.com's Game Review said that I made a mistake and I think it's completely wrong.

The first is pretty straightforward endgame. I have no idea why it's saying I should have moved Re4.

This second game is trickier, and the the explanation is crazy. When you look at the moves that it thinks will be made, it seems to me like they're assuming that I would just give away my Queen. Is there a legit reason that a player would do that that I'm missing?

I won by resignation, for what it's worth. I just can't understand why they think I would move Qxb5 under these circumstances though

Yeah, I thought moving my knight there was a damn good move

Okay, I see what you're saying. I agree. At that point I'm already -M8, so they assume everything will work out that way, so losing my queen helps me delay the mate. What about the other one? I don't think the pic came through initially, but I'm trying to attach it now.

The engine sees you have to give up your queen to avoid checkmate, so it says you're losing your queen... sure that doesn't make sense, but engines aren't known for being good at this.

Here's an example mate.
-

The engine may not have calculated all the way to mate after your move, so it says it's a mistake because it calculated a forced checkmate in a different line.

Since it only has a few seconds to analyze a whole game this is the result.

Or just download an engine for free, that way you can use on your own. Set it to show you the top few lines, and let it reach a decent depth before paying attention to its suggestions.

Either way, the lower the rating, the less engine analysis is able to teach you, since you have to see though the type of BS the OP is pointing out.

For this position, the computer thinks there is a more efficient way, so they label the check a blunder. Remember that the way a computer works is different from how a human mind works. Almost anyone on this site would know how to land a checkmate with a king and a rook, but the computer relies on a set of calculations to find the shortest possible path to checkmate. If the resulting check leads to an extension of checkmate by many moves, the computer might label it as a mistake.

But it is not "the engine" who makes the mistake here, it is the dumb chess.com script that tries to translate the engine's numbers into English sentences (and fails miserably).

Yeah, that's true.