Results and Rankings.

[Scott]

Hovering over the score to see which event it's from won't work in Internet Explorer 6 or earlier - it uses <abbr> tags, which aren't supported prior to IE7.

[roadrunner]

I think distracted has hit the nail on the head - if one group of people always run brown and another group always run blue, there's no way the points scored on the two courses will converge to something consistent. And if some of the "brown" runners run blue when they're unfit, injured or whatever, that just makes it worse. So, somewhat perversely, for the single ranking list to work, we not only need competitors to be able to choose to run any course, we need them to be less consistent in the choice they make!

[Scott]

But assuming a significant number of people continue to run the "recommended" course for their age class (which has been the case so far), you actually get quite a bit of cross-over of populations at different-sized (traditional cross-county long-distance) events.

To quote myself from a while age:

At a "medium-sized"* Long event, Short Blue will feature M65, W45, W50, M45S, M50S, M18S, M20S and W21S, all being ranked against one another.

At a "small" Long event, M65, W50 and M50S will now be on Green, scoring ranking points against M70-M85, W55-W85, M55S, M60S, W18S, W20S and W35S-W45S. Meanwhile, W45, M45S, M18S, M20S and W21S will now be on Blue, scoring ranking points against M50-M60, W18, W20, W35, W40, M35S and M45S. Of those, M50 and M21S would have been on Short Brown at a "medium-sized" event, sharing a course with M18, M20, M45 and W21.

And so on. And that's not even considering the different combinations of age-classes on courses at Middle, Urban and Sprint events, nor the daring people who don't actually run the same course every week.

*Where a "small event" has Brown, Blue and Green courses, and a "medium-sized event" has in addition Black, Short Brown, Short Blue, Short Green and Very Short Green.

As I don't really understand the maths, perhaps graeme or some other clever person can suggest whether this cross-over will actually be sufficient?

[graeme]

to get the single ranking list to stabilize it is essential that people run a variety of courses.

I don't know if enough mixing is going on. But I do know how it c/should be checked: repeatedly run all the results through the algorithm over and over again until it stabilises. When it was suggested to use quartiles instead of standard deviations, Mike Cumpstey ran this check, and it was stable. Possibly all the arbitrary cutoffs for "outliers" mess up the actual algorithm?

But I'm pretty sure the problems here are from the set up (or possibly coding the algorithm?): it seems clear that the spread of seeded scores was too small compared to what the algorithm is now producing. It's annoying me that I can't figure it out .

[Miner]

I have been following the discussion on the new ranking list with some interest as I used to look at the old ranking list sometimes to see if I was improving in some measurable way. Now all the inconsistencies and anomalies of the new list have been pointed out it has cured me of this pointless obsession with ranking points.

I realised it wasn't my place in the ranking list that I was interested in anyway, but my position within my peer group (however I might like to define that). How better to judge that than to travel around and compete against them. To broaden the peer group and make the comparison more meaningful simply travel further afield to include a diversity of terrain and larger events.

[Timsands]

Thank you to all those kind people who provided explanations as to why I am doing so well by the new ranking system in running Blue courses at small events (I moved up 38 places this week). My sympathies do however go out to Andy Hempstead and many of my M60 peers who competed at last week's National event. Oops am I still allowed to single out an age class? The results I know are given by course on the Brtish Orienteering website. Is this a deliberate strategy? Andy won the M60 course by a mile (6 mins+) but scored just 20 points more than my 1208 at last week's BADO level 2 event. Jim Prowting (and others) did much worse. Although Jim finishing second to Andy his ranking total was 50 points less than my score. I am totally unconvinced that once the system settles these discrepancies will reduce. Then I am an “oldie”.

[Gnitworp]

[The Loofa]

Timsands
Your comparision of points at the Midlands L1 and the BADO L2 is evidence that L1s should carry weighted points

[EddieH]

I don't agree with that Loofa - Graeme has convinced me

However I am convinced that the system is working really badly, and cannot see that scores will even out sensibly within a decade as it stands. Scott had a phenomonal run on Sunday in an area where inevitably a number of people make some big mistakes - you'd expect himto score really well, yet he just scrapes inside his previous best 6.

I don't see where there will be enough cross competition to balance the scores correctly.

Additionally there are some events which simply seem to give for more points than others, a statistical anomoly that will always be. When just 6 events count there will always be luck as to how many of these events you went to.

However as long as it is just treated as a bit of fun it doesn't matter too much.

Prepare to see me plummet as last year's scores disappear

[Scott]

It seems to me (without having done a thorough study) that everybody seems to be scoring lower in general this year than they did in 2009. Scotia's score from Sunday (and his similar score from Saturday) don't look amazing by 2009 standards, but they're pretty good compared to a lot of scores since the start of 2010.

Possibly this is a consequence of the way the algorithm was seeded? I'm not up to figuring out whether that's actually the case, but if it is then things might start to look a bit better as more of the 2009 results drop out.

[Spookster]

These were the Regional and National Events in March. The one's in red haven't uploaded the results to the British Orienteering system yet, so the associated ranking points are missing.

28 Mar 2010 DVO Regional
28 Mar 2010 South West WSX Regional
28 Mar 2010 EBOR Regional
28 Mar 2010 Scotland FVO Regional
27 Mar 2010 Scotland FVO Regional Middle
21 Mar 2010 Wales SWOC Regional
21 Mar 2010 CLOK Regional
21 Mar 2010 South Central BADO Regional
21 Mar 2010 East Midlands LEI National
20 Mar 2010 East Anglia WAOC Regional Urban Huntingdon
14 Mar 2010 Yorks & Humb CLARO Regional
14 Mar 2010 South West WIM Regional
14 Mar 2010 Wales SBOC Regional
14 Mar 2010 East Anglia SOS Regional
13 Mar 2010 UBOC BUCS Individual
13 Mar 2010 NATO Northern Nights (not a ranking event)
7 Mar 2010 South West SARUM Regional
7 Mar 2010 Scotland ELO Regional
7 Mar 2010 Wales POW Regional
6 Mar 2010 Scotland ESOC National British Night Championships
6 Mar 2010 Scotland INT Regional Scottish Sprint
6 Mar 2010 Wales POW Regional

[ba-ba]

...so Speaky won the JK, but has lost his #1 spot to OJ.
?

that said, I'm loving moving up 107 places even after 3 howlers in 2 races at the JK

[andypat]

Presumably OJ had a poorer result or two in there somewhere which got jettisoned? I went down over 100 places by not going to the JK - serves me right!

[ba-ba]

makes a bit more sense now!

[greywolf]

..so Speaky won the JK, but has lost his #1 spot to OJ. ?

...because scores from events over a year old are dropped...

http://www.britishorienteering.org.uk/e ... &course=7&