What is the baseline rating for someone who cannot play?

General discussions about ratings.
Ian Thompson
Posts: 3551
Joined: Wed Jul 02, 2008 4:31 pm
Location: Awbridge, Hampshire

Re: What is the baseline rating for someone who cannot play?

Post by Ian Thompson » Mon Jun 22, 2009 12:39 pm

Stewart Reuben wrote:That is a deeply flawed system. The player's start grade would depend on the average grade of his opponents and be -50 from that, or possibly -40.


The calculation of the start grade is given here: http://www.ecforum.org.uk/viewtopic.php?f=4&t=470.
Stewart Reuben wrote:It must be against extremely weak opponents yet who have a grade. Then what makes you think that a player playing random legal moves will lose all his games?
I don't think that. I think a player with no skill at all playing against other players with hardly any skill would be expected to score a little less than 50%. But that's not the question that was originally posed. The original question asked what the grade would be of someone with no skill at all who did lose all their games.

Stewart Reuben
Posts: 4542
Joined: Tue Apr 03, 2007 11:04 pm
Location: writer

Re: What is the baseline rating for someone who cannot play?

Post by Stewart Reuben » Mon Jun 22, 2009 12:58 pm

>"That is why in FIDE the player must score at least 1 point in their first game."

Er, should that be first tournament? I'm not complaining, just confused!< Kevin Thurlow

Well done. You spotted a typo. It should read in FIDE the player must score at least 1 point in their first event.
Not all chess events are tournaments.

Stewart Reuben

Sean Hewitt

Re: What is the baseline rating for someone who cannot play?

Post by Sean Hewitt » Mon Jun 22, 2009 2:15 pm

Ian Thompson wrote: The original question asked what the grade would be of someone with no skill at all who did lose all their games.
Such a player would be mathematically impossible to grade. As would a player who won all of his games.

Stewart Reuben
Posts: 4542
Joined: Tue Apr 03, 2007 11:04 pm
Location: writer

Re: What is the baseline rating for someone who cannot play?

Post by Stewart Reuben » Mon Jun 22, 2009 2:42 pm

I think you could use minus 50 or -750 internationally to determine the maximum grade/rating of that entity.
A computer that played random legal moves might not score 0. Indeed with enough games that would be a certainty.
Of course something that does not know how to play chess, for example a brick, cannot have a grade.
Stewart Reuben

Keith Arkell
Posts: 928
Joined: Tue Dec 30, 2008 2:10 am

Re: What is the baseline rating for someone who cannot play?

Post by Keith Arkell » Mon Jun 22, 2009 3:28 pm

The maths is beyond me,or at least doesn't interest me,but as an active GM playing against all sorts of opposition in a range of different types of events in the UK,from GM tournaments,International opens,strong team events and weaker team events,to strong weekend opens and a few ordinary weekend opens, I think I can imput something towards this debate.
I have only recently become active again on the UK scene,and have been allocated an adjusted grade of 229,down from 232.

It looks to me,judging by the 100+ games I have played in the last 12 months,that Roger's assessment is entirely accurate.For example,I expect to and do score significantly better than 75% against players around 204 or 207,and that is now - before lower graded players have received their bounties from the god of grading! Perhaps I am not the best example,because I am playing catch up to some extent after my period of relative inactivity.I think my new grade will be around 235 or 6,but Roger is entirely correct that these free points dished out lower down will simply filter up to us and we will indeed be 250+ or so within a few years.

By the way Roger,I think Mark Hebden has been far stronger than keene and Hartston for decades(if we forget about his current dip in form,which I'm sure is only temporary),and I don't think he was ever that much weaker than Miles and Nunn.When all were active together Mark scored reasonably well against these two,and wasn't hugely out FIDE rated by them.
Without realising it Mark will have been continuously improving ever since he began playing chess,and will continue to do so until age/health/motivation etc have their say.I say ''without realising it'' because we all improve together,with time,across the whole spectrum of chess strengths,so this improvement won't show in our results,unless we improve faster than the average;but it does show glaringly when we look at the relatively poor chess we played eg a decade ago when we had the same grade/rating.Because of this continuous rise in standards I think that by objective criteria Mark is probably stronger now than Nunn and Miles were a couple of decades ago.

E Michael White
Posts: 1420
Joined: Fri Jun 01, 2007 6:31 pm

Re: What is the baseline rating for someone who cannot play?

Post by E Michael White » Mon Jun 22, 2009 4:16 pm

I am glad to see this interesting and worthwhile topic has restarted and I am therefore moved to contribute.

If an inanimate object (IO) which knows how to play random chess, plays 10 games and loses them all, then the performance rating should probably be taken as the average opponents grade less 40 or 50, as Stewart has said. Ian Thompson points out the first ECF published grade (old basis) is likely to be calculated in the way specified by Howard Grist as the average opponents grade less 60. Both these views are correct which means that the lowest possible grade, if there is just one IO playing, is the average grades of the 10 lowest players in July 2008 less 60 points. Which is probably around -60.

Or is it ? If I understand the new starter grade process correctly then:-

If there is more than one IO playing then it is possible for an IOs first published grade to exceed the highest rated living player ! This would require a club for IOs be set up so that they all played at least 10 games against each other and just one IO has a result against a living player, which may be a loss. This setup would not be very different from some junior clubs. Under these conditions few would argue against IOs recording victories against other IOs.

If we analyse this in too great a detail there is a danger of this thread interiorating into a useful stabilising feature for the ECF grading system as the average grade of the lowest 10 (or even 30) graded players less 60 could be assumed constant from one year to the next so that ECF grading reviews could use this as a metrestick instead of assuming that the highest rated players have roughly stayed the same.