Rankings Consultation

[graeme]

I certainly have gone faster, but not over that distance

[graeme]

Yes, I have run faster, just not for that distance

and I ran a madmike-stuffing 3:59:59 on my debut in, um, 1982 (in 1981 I was too young )

55 2:30:17 2:30:17 Robert Baker

[geomorph]

I ran 2:29:58 in London 1988 - I know it's only 2 seconds - but if it had been the other way

Your and Rob Baker's results are brilliant. Never got that good at a large event (2:37:16 best at Seattle 1981). Did 2:29:01 at Seaside in Oregon in 1980, but we got blown round that day as 12 miles was with a strong following onshore, and by the time we turned round it had gone and most of the return was slight down hill (previous year the wind had done just the opposite and the winner struggled at 2:28).

[greywolf]

and for those of us not mad enough to trash our limbs with 26+ miles of tarmac, but would like to think that we could've, if we'd've been bothered, there's always http://www.fasterrunning.com/oldsite/calculator2.htm
...a 2.26 marathon at last hurrah!

[DJM]

Clive Coles wrote:
I am always amused by folk who seek to devise ranking systems that are "fair" to all. With the best will in the world different areas present different orienteering challenges. Some areas are a better test of orienteering technique than others. It is bound to be easier to accumulate ranking points at events where your age class leaders are not participating.

Can I correct one fundamental misconception about the (existing and) new scheme? It is absolutely not the case that rankings points depend on who turns up - the statistical algorithm used by both BOF and the IOF was designed specifically to compensate for the "who turns up factor" by comparing your time not only with the times of others on your course but also with the existing points of the rest of the field. The strength of the field is therefore taken into account.

The algorithm also compensates statistically for spreads of time caused by different terrain types meaning that it is no more easy to get points on fast TD4 land than on a physical TD5 course - the same standard of run should get you the same points.

[geomorph]

Clive Coles wrote:
I am always amused by folk who seek to devise ranking systems that are "fair" to all. With the best will in the world different areas present different orienteering challenges. Some areas are a better test of orienteering technique than others. It is bound to be easier to accumulate ranking points at events where your age class leaders are not participating.

Can I correct one fundamental misconception about the (existing and) new scheme? It is absolutely not the case that rankings points depend on who turns up - the statistical algorithm used by both BOF and the IOF was designed specifically to compensate for the "who turns up factor" by comparing your time not only with the times of others on your course but also with the existing points of the rest of the field. The strength of the field is therefore taken into account.

The algorithm also compensates statistically for spreads of time caused by different terrain types meaning that it is no more easy to get points on fast TD4 land than on a physical TD5 course - the same standard of run should get you the same points.

So statistically it is a level playing field. But it can do nothing about the terrain type specialist (do they exist?) who only attends certain events. But does that matter?

[EddieH]

DJM wrote

"Can I correct one fundamental misconception about the (existing and) new scheme? It is absolutely not the case that rankings points depend on who turns up - the statistical algorithm used by both BOF and the IOF was designed specifically to compensate for the "who turns up factor" by comparing your time not only with the times of others on your course but also with the existing points of the rest of the field. The strength of the field is therefore taken into account."

That is precisely why a serious ranking system should not include minor events, or should be weighted towards major events,) because they are events where many highly ranked competitors (who set the standard) have other agenda than a quality competitive run.

So if you care enough to want to get a high ranking, select a number of unimportant local ranking events and run them as though they are your major championship. Run to your normal technical standard and you are bound to get higher points than you could achieve at BOC, JK etc.

[LesS]

I have read the forum linked to on swat's piece and must say I have a certain sympathy with the comments posted. Given that the age classes appear to have been allocated to the correct colour of course, to change the courses/classes run by competitors after the race would appear to be a clear breach of the rules. If all the other events running under the new guidelines appear to have managed to submit the results correctly, it would seem a five minute job to resubmit the results and make everyone happy.

Whatever individuals think about Ranking systems, the current system is still being used this year for whoever wishes to use it, for whatever purpose. Since the consultation paper has only just been issued, there is no new single list system currently in place, proposals have not gone to the Board, nor have the membership been invited to adopt the new system so I actually find the final comment (at this time) rather condescending and a cop-out for not doing the work committed to when the event was registered as Regional.

Having O'ed for about 45 years, I must admit that I am travelling further to events now as I'm getting bored stiff with running in the same forests - scabby or good - over and over again. I obviously pick Regionals/Nationals on the basis that they should have better terrain and, and I get a competitive element as they will count for the National League/Rankings/Master's Cup. As DJM says, the presence or absence of other better or worse competitors is catered for by the weighting/current ranking of the other people you are racing - it's funny really, I thought everyone knew that - we all pick up different things.

More generally however, there are only a few people who are going to win races - the chances of the rest of us ever winning is virtually nil - even if no-one else turns up! Our competition, in effect, is with our contemporaries whatever the league we are running in - can I pull in a few points on those in front, and stay ahead of those below.

I must admit, I can't understand the need for a single list - if I wanted ranked against M21s, I'd run M21, if I wanted ranked against S runners, I'd run S, if women wanted ranked against the men, they'd run M classes, all teenagers - male or female - should be able to beat me, so finding myself some thousands down the single list is totally meaningless - even trying to oneself, let alone family members, friends and club colleagues isn't worth the bother.

I noted recently - can't remember if it was in one of the mags, on here or on BOF website - that the single list would be factored as per course lengths. Since the factoring has already been done at the event - adjusting for age and sex - would a single list that did not factor actually produce a better comparative picture highlighting the best pound-for-pound orienteer?

[madmike]

I ran 2:29:58 in London 1988 - I know it's only 2 seconds - but if it had been the other way

Your and Rob Baker's results are brilliant. Never got that good at a large event (2:37:16 best at Seattle 1981). Did 2:29:01 at Seaside in Oregon in 1980, but we got blown round that day as 12 miles was with a strong following onshore, and by the time we turned round it had gone and most of the return was slight down hill (previous year the wind had done just the opposite and the winner struggled at 2:28).

John Scholtens - an occasional RAFO runner (One time RAF Champion) - managed sub-2:20 for the USMC marathon a few years ago (and he was the wrong side of 40 at the time). Mind you he's rubbish on technical areas, even getting beaten by me.

[Scott]

I noted recently - can't remember if it was in one of the mags, on here or on BOF website - that the single list would be factored as per course lengths.

To clarify, as quite a few people seem to think that this is the case: the algorithm in the proposal attached to the consultation document features absolutely no form of "factoring" for course length.

[SeanC]

Eddie, would your point be addressed by a higher weighting for major events (one of the questions in the discussion document)?

There's lots of bias that must be impossible to remove. Those living in the main orienteering areas, those with the time and money to travel to events, those who only do orienteering and not other sports etc etc will be at an advantage. The list can't be a perfect measure of orienteering ability.

However if you are improving, so long as you attend enough events you should go up the list. This gives ordinary orienteers attainable targets, a reason to train, improve their technique and to turn up. We cant have PB's in orienteering so we need this sort of thing. Increasing the number of ranking events increases the number of people with this incentive. It's got to be a good thing.

That said I will probably be more interested in making the top 10 at the Kent Night Cup. No standard deviations needed as everyone runs the same course, and I'm competing against people I know.

Judging by the document you will be able to filter the single list based on criteria such as age, sex, club, region to make whatever list you want. Assuming there are enough criteria we can all find ourselves near the top of some list.

[Oldman]

I must admit, I can't understand the need for a single list - if I wanted ranked against M21s, I'd run M21, if I wanted ranked against S runners, I'd run S, if women wanted ranked against the men, they'd run M classes, all teenagers - male or female - should be able to beat me, so finding myself some thousands down the single list is totally meaningless - even trying to oneself, let alone family members, friends and club colleagues isn't worth the bother.

This is not my specialist subject, but I think you've missed several points here.

1. I doubt many people will want to see how they stand on the single list: rather, they will filter the list by age class and perhaps region.
2. I would assume that the appearance of everyone on a single list means that if two M45s enter different courses at the same event, they will still score ranking points as appropriate to their own performance and the prior standing and peformance of those they competed against. This will allow you to score ranking points even when running out of class.
3. The single list means that any "small entry" provisions in the ranking algorithm are less often required, since all entrants on the course count towards the total entrants, rather than just those sharing the same age class.

I don't especially warm to ranking lists myself but they do have a place for those more competitively minded, and also, importantly, for newcomers who are beginning to wish to monitor whether they are improving or not.

I hope all the contributors are feeding into the WG, although I suspect the WG will also get to see all contributions here and comment (as DJM did) where necessary to clarify or correct anything said.

[graeme]

I hope all the contributors are feeding into the WG, .

Fear not, the WG is watching us...
Upthread I posted about how to deal with mharky's problem of excess points for winning SOLs. It took about an hour for the WG to contact me asking for chapter and verse.

[The Loofa]

That is precisely why a serious ranking system should not include minor events, or should be weighted towards major events,) because they are events where many highly ranked competitors (who set the standard) have other agenda than a quality competitive run.

So if you care enough to want to get a high ranking, select a number of unimportant local ranking events and run them as though they are your major championship. Run to your normal technical standard and you are bound to get higher points than you could achieve at BOC, JK etc.

Well done Eddie H. You've got in spot on.
The basis for the maths behind the ranking list is that everyone tries equally hard each time they run - which patently can't be true. Surely most would try harder at BOC and JK and if everyone runs 5% faster because its the JK, you've got to run 5% faster too to get the same number of points, and if you don't you'll get 5% less points. Conversely at a colour-coded/district you get people going for a jog/ suffering a hangover/ testing an unprcatised technique/ recovering from injury and generally not many people trying that hard. So as Eddie says make one of them your top race and you'll probably get lots of points.

So the ranking system should exclude minor events AND have weighting on the most important races.

[The Loofa]

That is precisely why a serious ranking system should not include minor events, or should be weighted towards major events,) because they are events where many highly ranked competitors (who set the standard) have other agenda than a quality competitive run.

So if you care enough to want to get a high ranking, select a number of unimportant local ranking events and run them as though they are your major championship. Run to your normal technical standard and you are bound to get higher points than you could achieve at BOC, JK etc.

Well done Eddie H. You've got in spot on.
The basis for the maths behind the ranking list is that everyone tries equally hard each time they run - which patently can't be true. Surely most would try harder at BOC and JK and if everyone runs 5% faster because its the JK, you've got to run 5% faster too to get the same number of points, and if you don't you'll get 5% less points. Conversely at a colour-coded/district you get people going for a jog/ suffering a hangover/ testing an unprcatised technique/ recovering from injury and generally not many people trying that hard. So as Eddie says make one of them your top race and you'll probably get lots of points.

So the ranking system should exclude minor events AND have weighting on the most important races.