Progress and rating: puzzles ELO v's rapid ELO v's BOT's

Hey All - apologies about the weird title for this thread!

Essentially, I am curious about a few things that maybe some could enlighten me on pls.  Though I have been able to play chess for many years, I haven't played much at all and only in recent years have I joined chess dot com.  I decided earlier this year, rather than just to have a chess dot com account and do nothing with it, that I would make more effort and at least try to play some games to see what ELO I could potentially attain.  I consider myself pretty much a beginner, in my elder years (61) so no real potential for significant improvement, and over the last month or so in my few games I have managed to get to a rapid ELO rating of 717 - still very low for sure, but slowly improving.  Interestingly, my puzzle rating is 1499, and I enjoy attempting to solve puzzles a lot.

Here's my query: when practicing and learning, I play a bot rated at say 700, I win easily with total control of the game, when I play a bot rated 1500, I usually win - though maybe not as easily, however, if I were to play a player rated at 1500, I would be totally annihilated, and I mean totally.  Why is there such a difference in the playing level of bots rated at 1500 compared to people with a similar ELO?  What then is the point of a rated bot if it doesn't accurately represent the equivalent ELO of a player?  Surely, they should be somewhat on par with each other, yes?

Also - is there any correlation in puzzle ELO rating v's playing ELO?  Given I have a puzzle rating of 1499, is it inferred that I could potentially attain a player rapid ELO rating of somewhere around that level... or is that simply pie in the sky?

Just to add, I have no expectations of what ELO I might get to,  I guess I'd like to get to 1000, but not sure if I am capable.  I am curious about the relative ELO's of players v's bot's v's puzzles etc.    Thoughts...?

John 

I couldn't tell you in regards to bot elo. It is very misleading.

Puzzle rating does not related to a players chess rating. It's completely separate. I heard someone say once that you should try for a puzzle rating 1000 points higher than your chess rating. Whether or not this is true, I don't know. I'm around 1200 elo rapid and have a puzzle score of 2500. The only take away is that one year ago the best I could do was 1700. So even if it doesn't match up with the elo of my chess games, I can still measure it and see improvement in my understanding of tactics and positions.

I just want to say, even though you're in your 60s, that doesn't mean you don't have potential. Maybe becoming a titled player is unrealistic, but 1500-2000 elo? That's probably possible with study and dedication.

Humans create emotional pressure — blitz attacks, confusing sacrifices, or sharp endgames.

Bots feel sterile, almost calm — and that’s not what most games feel like.

Bots are overrated. I would focus on playing humans. Puzzle ratings are also distorted, and I would not rely on them as an estimate. Simply play against humans and improve! Good luck!

This AI-generated answer is terrible. This is why you can't trust AI - they just say something even if they know nothing, and the result is utter BS like this one. Not a word of it has any connection to reality.

The real reason, obviously, is what everybody else gave. Chess.com't bots are overrated.

I mean the AI response is not wrong that is a correct difference between humans and bots but it has nothing to do with what the original question was.

The irony of "bots feel sterile" when this came out of chatgpt and it's about the most sterile statement one could make.