Forums

Opponent has ridiculous best move percentage

Sort:
SneakyWalrus
 
I was black in this game, got beat fair and square. I noticed something on the computer analysis though, this guy (rated 1100-1200) made the best move over 70% of the time. This is by far the highest I have ever seen with someone in my rating range. Any idea why? Did he get lucky? Just easy moves for him to see? Thanks for any help. Just trying to understand how he saw the best move so often.
urk
White's play looks pretty bad to my eyes. Is that what you want to hear?
I don't see anything suspicious.
SilentKnighte5

"I got beat fair and square, except it seemed suspicious".

Diakonia

" I noticed something on the computer analysis though, this guy (rated 1100-1200) made the best move over 70% of the time."

And yet you lasted 54 moves.

ModestAndPolite

70% best moves means 30% of his moves were inaccurate, mistakes or blunders.  That is a lot of inferior moves.  His moves look like human moves and not machine moves. As Urk said, White did not play exceptionally well and there is nothing suspicious in his play.

AussieMatey

White blundered a pawn for nothing on move 7. No need to go any forther - he's definitely not cheating.

shcherbak

It is not very complex game and every time you have made a blunder your opponent do not have much choice, but to make good move. If you want to benefit from analysis, start from understanding own mistakes. Accusing, or insinuating cheating in public is also kind of blunder.  

ArtNJ

I imagine you think you are being subtle, but a mod may well still reprimand you for naming your opponent and basically accusing them of cheating.  Naming your opponent in the form and being critical is a big no no.  If you are really just curious and did not intend to imply your opponent was cheating, you should still have removed his name from the PGN given the nature of your question.

Just FYI, people that think someone else is cheating are almost always wrong.  Mostly because a regular cheater would not be rated nearly as low as they are.  And, also, as already pointed out, because they are misinterpreting the significance of the computer's move evaluations.  Detection of cheating is a *very* complex subject and not something most, and especially beginners, should concern themselves with.  My impression is that cheating mostly goes on in correspondence, and is fairly rare even there, at least at levels below master.  

Zerotgb

I have a friend of mine who used to play chess against other players here. He cheated. He opened another match against a high-level computer and imitated the human opponent's moves against the machine. That way, the only thing he needed to do was imitate the computer plays to win against the human player. Maybe that's your case.

ModestAndPolite
Zerotgb wrote:

I have a friend of mine who used to play chess against other players here. He cheated. He opened another match against a high-level computer and imitated the human opponent's moves against the machine. That way, the only thing he needed to do was imitate the computer plays to win against the human player. Maybe that's your case.

 

That is theoretically possible, but in this case, as has already been pointed out, a computer would be unlikely to play so badly. Even on low levels the mistakes computers make are not like human mistakes.  In the game in question both players make very human mistakes.

AIM-AceMove

At that level it's are very common 2 things.

1. Everyone knows what i am talking ccccc

2. It's its very common so called "sandbaggers" - Players with higher skills, but with low rating, playing at low level just to play easy games and to win more.

Just 2 months ago i got beaten by 1150 rated... positional... and i was playing serious.

 

When i started playing chess again in my first months i went around close to 1500. Then i decided to delete it and to start new and fresh. To my surprise i was nowhere near my old stats - but i was down to 1150 - playing a lot stronger players than their rating indicates... Keep in mind that there are thousands of players with 10 000 or 30 000 played games. They can play a lot stronger than normal 1200 rated with lets say 1 000 games.

AIM-AceMove
ArtNJ wrote:

I imagine you think you are being subtle, but a mod may well still reprimand you for naming your opponent and basically accusing them of cheating.  Naming your opponent in the form and being critical is a big no no.  If you are really just curious and did not intend to imply your opponent was cheating, you should still have removed his name from the PGN given the nature of your question.

Just FYI, people that think someone else is cheating are almost always wrong.  Mostly because a regular cheater would not be rated nearly as low as they are.  And, also, as already pointed out, because they are misinterpreting the significance of the computer's move evaluations.  Detection of cheating is a *very* complex subject and not something most, and especially beginners, should concern themselves with.  My impression is that cheating mostly goes on in correspondence, and is fairly rare even there, at least at levels below master.  

 Not completely true really... bot not the place to talk about this.

blueemu

I don't see anything suspicious about White's play in that game. It was pretty bad, to be honest.

BronsteinPawn

Are you really saying the guy cheated when he played 3.e5?

Learn from your losses instead of saying to yourself he cheated just so you can sleep at night.

BronsteinPawn

Maybe if you knew the basics about chess you could have tried to punish your opponent for overextending your pawns with 5...c5 instead of the semi-dubious 5...Nc6

But hey, he hung a pawn later on, and you hung a fork! So everything is good!

SneakyWalrus

I mean obviously he didn't cheat, he made his moves pretty quickly and he did blunder a pawn at a point. I should've worded that better. However, my "best move rate" was around 40%, so it's just weird that I lasted so long, is it not? Is best move percentage an accurate indicator of anything? 

urk
You should have worded it better.
blueemu
SneakyWalrus wrote:

Is best move percentage an accurate indicator of anything? 

Only in tactical positions. In quiescent positions (no tactics), engine evaluations are basically worthless.

bryanb1980
I hover between like 1050 and 1100, pretty low rating. I'm not very good obviously, I sorta just started trying to learn again recently. But I've had games in the last week or so where I've played well and probably hit over 70% "best moves" or whatever. I usually do the quick analysis thing from the site and I'll have games where there 0-2 inaccuracies or mistakes. Usually it's when my opponent plays pretty bad from the get go and the right moves are pretty hard to miss. Doesn't happen a lot but it does happen so I wouldn't think much of it.
macer75
blueemu wrote:
SneakyWalrus wrote:

Is best move percentage an accurate indicator of anything? 

Only in tactical positions. In quiescent positions (no tactics), engine evaluations are basically worthless.

It also depends on how bad the moves that weren't accurate are.

This forum topic has been locked