<This tournament is just one case showing how unwarranted Frogbert's position was.>

Of course, but as <Appaz> stated:

<....(T)he finish was not convincing at all. It's not a proof (of) psychological weakness, but certainly weakens the claim of the contrary.>

Can the ELO rating system allow, in some circumstances, for a rough ELO rating to be infered for one player, who is unrated, from the results of a match with another player, who has an ELO rating.

For example, (from a recent actual circumstance), players X and Y have an 18 game match at classical time controls, where the results in favour of player X are +10 -4 =4. If player Y is ELO 1550, can a rough estimate be inferred from these results for ELO of player X?

---

<I believe I said that if we paid attention to the ELO rating formula, we could end <at least one silly debate>, namely whether or not ratings can be used to <predict> outcomes with some degree probability, something that was being denied (but not by you I don't believe).>

pbercker, what's silly is pointing to the Elo calculation formula in order to prove that the rating system was <created> with the <intended purpose> of being able to predict outcomes. And yet, this was the "silly debate" in question, if you would've paid close attention back then. I certainly <did deny> that the purpose of the rating system is to predict the outcome of chess games. You seemed (and seem) to think that the Elo calculation formula provided some argument in this debate, and you even labeled the idea of having said debate "silly".

I can explain to you why you were wrong (and still are wrong) if you *want* me to. But you think googling this or that makes you able to settle such a debate (while labeling it "silly"). Today I offered my player page to continue our discussion, but you insisted on continuing here. I'll make the same offer again: take it somewhere where we don't bother those who don't give a damn about this off topic discussion, and it *will* be possible for you to learn something. But first you need to realize that I can teach you something.

<You're quite right ... my apologies ... I allowed myself to get sucked into another useless argument with Frogbert>

I won't bother to "argue" with you. I've offered to explain to you why your ignorance of a topic I understand much better than you was annoying me back then, and you still seem to insist on being annoying instead of accepting my offer to explain why you were wrong then (which you still are). You make wrong assumptions all the time, like when you assumed that I put you on ignore back then. Stop making those wrong assumptions based on limited knowledge - it will hurt you in a context that <does> matter some day, unlike this one.

<Your evident lack of manners doesn't bother me too much, as I can overlook it.>

No, you can't. You've just proven multiple times that instead of ignoring my seemingly bad manners and accepting my offer to talk about this subject without any "audience" in a quiet place, you would rather make personal slights based on more of your wrong assumptions. Which kind of proves that I was right in the first place when I just left you and AgentRgent to your misunderstandings and misconceptions back then.

Yes, I can be arrogant, and yes I can easily give up on people who don't deserve being informed when they're misinformed, like you are if you write and believe the following:

<I said that if we paid attention to the ELO rating formula, we could end <at least one silly debate>>

No, it just proves that you've misunderstood the purpose of the Elo rating formula. You've googled some information and not been able to wrap your mind around what it actually means. Similarly, referring to a Chessbase article proves nothing, because Chessbase has posted loads and loads of provably false information ("garbage" in plain english) about ratings, inflation and so on for years and years now. Neither Chessbase nor those who write for them are infallible in any way, and regarding the topics I'm interested in (read: chess ratings) they've provided lots and lots of <opinions>, some well-founded and others utterly unfounded, from various authors and sources, which taken together provides a totally inconsistent and incoherent picture of what ratings are and can/should be used for.

That's not really any fault of Chessbase - it's a good thing that they let people with different views and opinions voice their opinions, but it totally destroys the idea that any single view/opinion posted on Chessbase on this topic bears any importance (beyond that of being someone's opinion) or that it can be taken as "proof" of anything. Chessbase is simply a commercial player in the chess market - they aren't truth seekers or researchers, and neither are most of their external contributors.

Short summary: you're welcome to respond to this post on my player page, quoting/responding to whatever in this post you like, but I <will not> continue this exchange here, on the Norway Chess tournament page.

And again, my player page is here: Hans Arild Runde

The <purpose> and the <intended purpose> need not be the same.

That its <use> is fundamentally <predictive> with some computed probability is absolutely unarguable and was indisputably at least <part> of its <intended purpose> and part of it <current> purpose.

That <ratings> is <correlated> with <intrinsic strength> is well known. This is because <intrinsic strength> is causally related to <performance> and <ratings> are a record of that performance. As such, ratings are an <indirect> measure of <strength> and generally <intended> as such.

I generally disagree with at least 90% of what AGENTRGENT says, and I am predisposed to dislike his style of arguing as it too often veers towards nonsense and illogic. However, in a rare moment of logical lucidity he managed to summarize the entire <silly> debate quite well:

<AgentRgent: <frogbert: your description of ratings as mainly a predictive tool was and is completely wrong> I said that Elo's intention was to design a system that could reasonably predict who would win between two players. Ratings were designed to be a predictive tool using past performance as the guide. I agreed with you then and now that their failure to be perfectly accurate is not a "failure" of the system. I suppose that the main difference between us is the idea that Rating=Strength. Which honestly I don't think is really a difference of thought more than a slight difference in how we express ourselves.>

Not only is he correct, but he admirably ended by giving you the benefit of the doubt, but which only belatedly I see you did not (and do not) deserve.

Just stopped by your page to give you a lovely quote that you can use again and again :)