The way any chess engine works is that if it gives you a 20 ply variation, the first move was searched 20 ply, the second one 19 ply, etc, up to the last move which will be a 1 ply blunder. There's nothing specific to chess.com about that - it's just how engines work. The low quality of the moves at the end of the variation has no bearing on the accuracy of the moves at the start. If it had, engines would be weak, but they obviously aren't.
Membership's deep analysis is gone?
Sort:
Besides that the new system is NOT accurate! I did a retry. It supplied a move variation based on my new move. It appeared under the green bar that appears when one gets a blunder right. It populates a moves variation starting from one's new correct move. I clicked at the end of this variation and I was immediately taken to the analysis section where this new variation was now in the game move list. So I decided to move through each of these new moves to understand the logic of the moves. As I moved through each move the engine was re evaluating these moves. AND IT GOES AND FINDS 2 BLUNDERS IN IT'S OWN RECOMMENDED MOVE VARIATION!! Besides these two shockers were 3 inaccuracies as well. In its own recommendation! So right now I have serious concerns about the accuracy of the new analysis engine and system if it can dish up a suggested variation that one assumes consists of "best"moves but actually contains blunders and mistakes! I was quite taken aback to say the least.
I have reported it now to customer support and I await their feedback.
@big_data_is_cool the above quote is from another post thread on the new engine that I wrote. I am including it as I noticed you picked up that the new analysis appears to be inaccurate versus an analysis you have run comparing the two. If I may ask...what program did you use to generate your comparison analysis? Also...just for your fyi...Stockfish is at version 10 now (not that there is anything wrong with Stockfish 8...just Stockfish 10 is improved and stronger!)
I am awaiting feedback from customer support on the major inaccuracies I found with the new engine when I started using it. Right now based on the hard evidence I have I personally don't trust the new analysis's accuracy levels.
Did you ever hear back?