Opponent's Average Rating

during my first 100 to 150 games (or so) of online chess, i never really paid much attention to what my opponents' average opponent ratings were.  i've started to look at this a bit more.  i'm wondering how telling you all think this stat is.  lately, i've taken to playing higher rated player and in the past couple of weeks my opponent's average rating has climbed from the mid 1400s to the mid 1500s.  this makes me happy as i feel like i'm playing a higher quality opponent.  that being said, i've also started losing more than i was a few weeks ago.  losing is painful when you're putting a lot of effort into games, but it's not so bad to lose 5 or 6 games to 1800s with the odd win here or there, if you feel like you're getting better.

anyway, i started looking at some of my opponents that i either didn't really buy their rating as matching their play or that i was able to beat pretty well.  not across the board mind you, but some of them had really low O'sAR (easier to type).  one high 1700 i played had an OAR of like 1400 or something.  i'm thinking--lame.  step it up pal.  it's like Man United sweeping through the 1rst division.  anyway, i'm wondering if any of you look at this.  how telling do you think it is?  how important is it to play people at or above your own rating?  is it considered by anyone else here--kinda lame to always play down to be able to rack up wins?  i suppose it's only opinion but i hope to generate some discussion.

I think the O'sAR is often more informative than the actual rating.

one thing i was kinda thinking about is this--i suppose as a player is getting more wins and moving up the ladder, so to speak, they will gain a higher rating.  if it stands to reason that players often play other players with similar ratings as themselves, then many players would have OARs lower than their own ratings.  if i'm a mid 1400s player then i would play a lot of other 1400s players.  as i find myself winning these games i will turn into a 1500s player but many of my games would been played against 1400s.  i suppose this is one reason why OARs would be lower than a player's own.

plus, it's an average...

richie and oprah, your rating is a good example.  i looked at your best win opponent's OAR and as a 2300 his OAR is mid 1800s--the same as (maybe a hair lower than) yours and you're a 2000 player.  i find that pretty weak and week of him.

beech and beach.,.,.  ya know?

anyway, not trying to call anyone out but that's a hugely telling part of the rating system.  i'm not sure exactly what it tells but it tells something.

If everybody is fairly rated, it shouldn't really tell you much.

In theory.

what do you mean "fairly rated"? 

rated correctly?  or that the rating they receive is given to them in a manner that matches the way everyone else is receives their rating?

Another thing to consider is that games against lower rated players may finish more quickly, so even if someone keeps a fairly balanced set of games going at any given time, that person will play more people with lower ratings.

Plus, someone might play a single game with an opponent who is rated many hundreds of points above or below, and that could shift the average disproportionately. Maybe a median opponent rating measure would be more useful.

I don't know. I usually set up my challenges for a range from my rating to a few hundred points above. But my OAR was still way lower than my actual rating for a while after my rating went up significantly from a few wins - but then I had a string of losses and my OAR is now higher than my rating. Maybe I see more variability because I'm a slow player so my game count is low.

I place more emphasis on the stat average rating when winning.  I like to know the ratings of the players they tend to beat.  If someone has an OAR of 1500 but are only beating 1000s I tend to discount that player as being someone who is choosing tournaments and games carefully to inflate the rating.

I think you would need to know the exact forumla to in order to assess how useful the stat is (are oppenant provisional ratings included, etc):

If it is just a simple case of adding the ratings of all oppenants and dividing by a the number of games you might end up with a slightly misleading statistic.

More useful to know, might just be the mode rating of all complete games, rounded to the nearest 100.

I disagree, I started playing chess a year and a half ago and my rating was around 1000. of course, it wasn't easy for me to get to where I am today which is 1974. I had to play people of many different ratings so my averege is 1500.if I ever do play a lower rated opponent right now it would be a friend I met when I was lower rated or a tournament I'm in, or for tutoring. you will see many games called "practice", "mentoring" etc., but look at my games right now you will see me playing many high rated oponents.

I prefer playing stronger players but I accept all challenges, rating doesnt matter

I agree -- anecdotally I've found this to be the case in practice, so it's one of the key stats that I do pay attention to -- the others being the actual rating (which in combination with the oar give me an indication of strength), the time per move and the timeout percentage (which together give me an indication of speed of play and general reliability).

I have a friend here who is 1900-2000+ with an average opponent of 1800+ and a best win of 2300+, My rating is about the same as his on average but with a 1677 average opponent and a 2000+ win that was a timed out game. Yet i have only lost once out of about 7 games to my friend. How does that equate to u? I have no idea exactly y i am able to do win so much since he is so very good himself.

Not to mention u don't know if the average rating of the player dropped because of time outs from opponents who would have been much higher rated when playing and things of that nature. I think average opponents ratings is a guide but u cant always assume they are weaker players by that. Remember they still have to win the majority of games to be at the rating their at....

if anyone thinks that my rating is high only because I play lower rated opponents then I challenge them to challenge me I'm confident in my ability against anyone less then 2000

i wish that chess.com would rate us with time-out wins and without time-out wins.  i think they should count, but it would be nice to see a rating without.

In Chess winning isn't really everything. Sometimes the Greater player need not be better than the weaker ones. Losing games are just as important as winning ones.

The only goal is to discover how to improve your play and you just realise this as you get to find playing stronger players just as easy as the weaker ones! So play on and let the Battle Begin!!

-P-

my opponents avarage rating is about 60 points higher than my own, which may explain why I have so many losses.

yeah very zen and all, but that's not really what i'm talking about.  i think it's a fairly universal belief in chess that learning from your losses and mistakes is a great source for becoming a better player.  that's not what we're talking about.  i want to talk about opponent's average rating and how each player reads that stat.  so Phyyrrus, how do you feel about your OAR or other's OARs?  hehe

I agree with you about the fact that if your average opponent is a few hundred points lower it is kind of "lame." My average opponent rating was mid 1400s and I was about 1900, but then I started looking at that stat and I didn't like such a wide range between the 2 so I started playing higher rated opponents. I have played a lot of games so raising that stat is not easy, if I play someone about 500 points higher than me it may raise that stat by 1 point now, it usually takes quite a few games to do so. So for awhile now I've only played people 1800+ and the average opponent rating is in mid 1600s now. Now that I've started playing so many tournaments it will likely drop since draws are random and I can't go only accepting opponents rated 1900+ in tournaments.

But yes I think it indicates well whether players are playing hard games and have earned their rating, or if their playing much lower rated opponents to get an inflated rating.

I have to admit it matters to me what my average opponent stats is, maybe more then it really should. The trouble is sometimes i get games time out on me which drops their average rating down 200 or more points, and i had good games going too. You also have to understand that often just as in OTB tournaments some players ratings are still on the rise, even though their average opponents ratings maybe are a lot lower as they are climbing so fast it hasn't caught up yet.

I believe i deserve the rating i have, maybe 100 points higher but i am not sure as i am still testing myself against opponents. The fun is in the challenge i agree but sometimes its important for ur lower rated opponents to be able to improve themselves by playing stronger players like u too.