ERATIC INCONSISTENCY IN GAME REVIEWS

I play almost exclusively against bots these days. Has anyone ever noticed that for no rhyme or reason, they get a very low game rating in a game where they played exceptionally well, employed superb winning tactical combinations, and made several great, if not brilliant moves, winning the game. Whereas, in other games where I have played well, but not exceptionally well, I received a much much higher game rating than I did in some games where I played much better?

I, too, play bots exclusively. I'm playing about a 1400 level and review most of my games. I've noticed the review seems to be based exclusively on the last move made without regard to where the play might be going. Like if I set a trap and it works the decoy move gets a bad comment. The review doesn't take into account what's about to happen - until it happens.

But I have one question maybe you can answer about the bots. When I'm really in the zone I can play pretty well against the 1400 bots, but on a bad day I can fall back to the 1100 bots (I do love playing Aziz) and get beat game after game against different bots at the lower level. Then another day things perk up and I do very well. My question is, I know I can be "in the zone" sometimes, and sometimes not. But do the bots get "in a zone" for a while and stomp me more than normal, or do they play very consistently and it's all me? I almost hope the answer is that the bots get into the "zone" sometimes just like playing human players, but I'm expecting the answer is that it's all me.

No, I don't believe it is all you. Sometimes the bots -- even the relatively lower rated ones -- play really wickedly, like virtually at a master level. Other times they don't play that impressively at all. I'm not sure what the reason is for this inconsistency. I do tend to have a high percentage of wins if I get them into the end game. I think it is because they are super well-programmed with openings and even early middle games, but when they are faced with a new situation they haven't seen before, which is typical in end games, they get confused and make some really bad moves. What has helped me improve is to play a bot at or close to my level a lot, and once I start winning the majority of games or at least as many as I lose, I force myself to play the next higher level bot every so often. This usually results in me getting killed at first, but after a while I start drawing and even winning some of the games. For the past several weeks I was playing this Bobcat Bot and eventually started winning the vast majority of the games. Today I moverd up to the 1800 Bot and expected to get slaughtered, but I made a draw and even got a 1900 game rating, while the Bot got 1850. I have found that It really starts to build confidence and increase skill to challenge yourself by playing stiffer competition.

Good to hear about the bot skill level. That's more like playing a person. When I used to play a friend years ago and made a dumb mistake I knew that if I stayed in the game long enough there was a good chance he'd make an equally dumb mistake. That's what made it fun. (and the more we'd drink the more fun it would be!)

agreed. I do basically the same thing but at my level. The bots usually play really bad at an end game, even sacrificing their pieces needlessly and running the pawns up to my pawns. It seems to just give up and go for a stalemate. Although I get sloppy too at the very end and have let him trick me into a dumb stalemate move cuz I didn't look close enough to see his king was trapped. Early on I found my two worst weaknesses were my sloppy ending play and what I call 'look before you leap' mistakes where I analyze the bd to death and then make a dumb move, which I saw at some point during the analysis, and lose a piece.

you're absolutely right about playing above your level. I always learn more by losing than by winning. But I need to win enough so not to get totally discouraged and give up chess. Finding that balance is key.

I know the feeling of wanting to give up Chess completely. I 've had some losing steaks that had me pretty discouraged. But I stuck with it because I love the game and it can be relaxing and fun and sharpens the wits. (When I drive I see it as a big chessboard and look in all directions so I don't "blunder" or get blind-sided by a sniper "bishop" or something, lol. I really hope you stick with it. I highly recommend focusing on just one opening for white and one for black and get to know those really well. Video lessons are really helpful for that, and after using them a lot in games you become aware of the pitfalls and traps and what works and what doesn't. Also, doing some puzzles on a consistent basis really fosters improvement.

You're better off doing your own analysis if you want to improve.

Don't worry, I'm not thinking about quitting chess. I absolutely love my Chessup board. I usually do the analysis for most of my bot games, but I don't always agree with it's comments. I don't know if the analysis is so far above my level that i don't see it, or if it only looks at that instant in the game and doesn't get where i'm going. I guess I'm a little old-school in that I prefer to learn by getting beat up a lot until I figure it out instead of relying on somebody else's theories that I probably don't fully understand. Either way, it's all about enjoying the learning process... the journey, for me anyway. Lots of people want to BE a rock star (or millionaire, or whatever) but few people want to BECOME one. To go thru all that it takes to get there.

Here is a great article about Reviews: https://www.chess.com/blog/Duckfest/game-review-common-confusion

Thanks for the article David, and glad to hear you're not thinking about quitting noluck. I do like to run through the computer's analysis because there are some excellent points, but other times when it has said I made a mistake, it was on a good or even great move, for example, one that won me a piece or that was part of a tactic that won me a piece. So there is the very inconsistency that led me to start this topic. The last few days I have played against the "Sliske" bot, whom the computer rates at 1800. The first game was a draw, the second and third I got slaughtered and the fourth I won in the end game and actually received a 2100 game rating - my highest ever. However, my high was short-lived; I then proceeded to lose the next 4 games against Sliske. He actually seemed to get far stronger than he had been, which is another point of confusion I mentioned previously in one of my replies. Go figure...

@dowaliby, we all go up and down and it has nothing to do with our rating or Reviews. It is because two humans are playing. If we let ANY rating or Review dictate our 'value' in chess, I think we hurt ourselves. I no longer use Review to decide how I did in a game. I've gotten a 91 when I KNOW I truly didn't deserve a 64; same with getting a Brilliant.

As @LieutenantFrankColumbo said, "You're better off doing your own analysis if you want to improve." Agreed! Do you own analysis before ever allowing yourself to look at Review so you're not influenced to believe your (Review's) press about you. Only after thorough self-analysis, then glance at Review and more accurately agree or disagree with it (my opinion only of course).

ADDED: I changed the 'paraphrased' to a proper quote like a good person would do. :-)

Great article. Hopefully as I review my future games I'll know when to put more (or less) weight to whatever move the review suggests.

Okay. Sometimes if I've had an exceedingly delicious checkmate or Queen take, I'll look quickly but try not to let it sink in. ROFLMAO!

@noluck987, agreed - it's one of our many tools that ChessDotCom has made available to us. So many rail on them but truly the user tools alone are very cool!