The FICS rating system

Here is another example of a reason not to care about it.

Creating: LinuxGuy (1718) VonSFdr (1986) rated standard 15 5
Challenge: VonSFdr (1968) LinuxGuy (1731) rated standard 15 5

My rating goes up a whopping 13 points for beating someone rated 270 points above mine, while his rating goes down a mysterious 18 points (I couldn’t even guess on how they calculate this stuff, but it seems like crap, to me, if you are the one trying to _maintain_ a certain rating).

Like I say, I couldn’t care less about my FICS rating, and sometimes I have simply let games go. I certainly don’t hyper-ventillate over the keyboard or nothing – only get really excited like that playing OTB.

Here’s the game score:
Caro-Kahn – Advanced var

Anyway, I just found it funny following Crafty’s point-score on some moves, since it’s usually a given when I win this easily that Crafty will be saying I played like crap. But after a minute or so, Crafty will suddenly start liking some of these moves that I played.

It’s interesting, Crafty says Black is winning with 18…Ne4 -.4, which I thought he would play, BTW. I post this game partly because Chesstiger does not like my e5, Ne2 variation that I made up. Well, Black got …f6 in like tiger says, and still lost. Does the variation suck for White? Who cares? If it can win at my level, that’s better than losing against the main-line caro, which I used to do consistently versus weaker opponents. I mean, perhaps five minutes from now I will change my mind about this variation, but it’s not like I’m losing sleep over it.

Incidentally, after 18…Ne4 19.Qe3 0-0 20.Rf3 RxR 21.gxR Ng3 22.Bh3, our fair-weathered Crafty friend, whom we’ve all come to know and trust (sarcasm), has changed it’s tune to .4 in _White’s_ favor.


12 thoughts on “The FICS rating system

  1. That’s ridiculous. If anything it ‘should’ be the other way around. Thanks for the link, though. 🙂

    Perhaps the logic is if you are playing a lot, you are probably playing like crap, which is definitely true in my case, but I think that’s partly because I get sick of the stunts that many players pull. By the time someone plays, you are already tired of waiting, then they start out slow, but speed up later, and by then I am bored out of my gord and have no patience left.

    By the time the Standard game gets interesting, I am already engrossed in something else (TV, reading, cleaning, etc) and make a simple blunder. But I don’t disconnect like quite a few do.

  2. Actually I think the system makes sense: the more frequently you’re playing, the more accurate your rating ought to be, and therefore it should change more gradually.
    Incidentally, you’ve nailed the reasons why I don’t play standard games online.

  3. I would have played 18…Ne4 but having said that i proberly would not have seen half the moves leading up to Ne4.

    Nice finish i also like the way you maintained the d and e pawns in the center,making the play on the flanks.

    I suppose you could say black should have castled at some stage.

  4. I would try to avoid exchanging of your nice light-colored bishop, but not sure if it’s possible, since after 17. … Rc8 your king can be stuck in the center. His main mistake was allowing penetration of your knight and queen on the queenside. This combination – Q+N is always brutal (remember my “Mate in 8”?).
    I just ran your game through Fritz, it kind of confirmed what I thought (also about Ne4):

    – 17. Bd3 with later Bxf5 exf5
    – instead of 18….Nxf1(+1.18) – Ne4 (-0.32)
    – allowing Nxb7 was a crucial mistake ( ~-7)
    – 27….Qe2 allows mate

    In the opening Fritz thinks you are a bit better (+0.3) until move 8, where it prefers h4 ( after c3 Black should play c4 with =).

  5. Hey Linux Guy!

    I am back from my work exile. Now I have the flu! I don’t worry about my rating at at all because it is much higher then it should be!! It seems things are weighted kind of weirdly. My rating is much higher than my OTB rating. And the OTB rating is the one that is a better barometer of where I am as a chess player.

  6. What this show about rating or “rating” system on fics, well its pretty easy : turbulences 200-400 points are something very common, with which system allowing selecting opponents has to live. FICS “rating” is uncomparable with FIDE (imagine Carlsen to have one day 2800+ and few days later under 2400 – sounds like nonsense true ?) ….

    Now question : which skills measure fics rating ? How big(in real seems low) is preciseness of measured values ?

    Sounds that right is this hypothesis : with accuracy X +-200 is X actual rating measured by fics. So if somebody has on fics in bughouse rating X, then his strenght (of whatever measured) is from somebody able lose with “himself” (=somebody with same rating) 0:10 to somebody able beat “himself” 10:0. So it saying almost nothing. Or we can say it is divide players into levels of 400 points groups – like 1000-1400,1400-1800,1800-2200,2200-2600 thats all, bigger accuracy is impossible with current rating.

    The another question is what the best measure and what the rating from viepoint of skills measure. Sure if rating should measure not good moves of team, but more human weakness and ability of premoves in 2 0 like time controlls, fine. This view of bughouse on fics as more or less arcade mouse game can be for some peoples fine. But it is not fine for anybody who likes exact rating system which works for FIDE and for somebody who likes measure good moves of team (human or computer or half computer doesnt matter). ….

  7. What this show about rating or “rating” system on fics, well its pretty easy : turbulences 200-400 points are something very common, with which system allowing selecting opponents has to live. FICS “rating” is uncomparable with FIDE (imagine Carlsen to have one day 2800+ and few days later under 2400 – sounds like nonsense true ?) ….

    Rating system on fics is better than FIDE (because of the RD). The turbulences come from the time which is fast and the little serious of players who don’t only play the games behind the computer.

  8. “Rating system on fics is better than FIDE (because of the RD)”
    well its nonsense : i can play OTB or with friends unrated or computer every day and my power dont decrease, but RD increase, so if I will not play on fics but will training every day outside fics the logic of glicko totally fail (no playing dont mean losing level of play skills) and changing thanks to possibility of selecting opponents into new best way how to fast increase rating : find opponent which dont like Your style, one You have always best performance against (not like in fide where random will be selected for you or atleast more various) and play only when Your RD is big, then You will gain most of rating points from less games, easy isnt it ? glicko simply say : if you wish have big rating , You have to play as low number of games as possible but against opponents You are able beat with high probability (its better play +5 against much weaker opp when you have big rd than +1 against same opp when you have low rd) …

    Glicko is much worse than FIDE because is based on bad=unreal assumptions, that who dont play losing skills fast.

  9. I think the rating system is flawed…big time. Am sick of playing 1100 players with 1500 ratings. This happens because they play much higher rated players all the time! They need different categories or something……I truly think this system sucks :O)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s