British Sprints
Moderators: [nope] cartel, team nopesport
66 posts
• Page 4 of 5 • 1, 2, 3, 4, 5
Re: British Sprints
The ranking software doesn't seem to mind if someone runs two different courses at the same event. So for the recent SE Sprints I just combined the round 1 and round 2 results into a single CSV file and submitted. No need for any dummy event, just some minor editing of course numbers.
- NeilC
- addict
- Posts: 1332
- Joined: Fri Dec 17, 2004 9:03 am
- Location: SE
Re: British Sprints
NeilC wrote:The ranking software doesn't seem to mind if someone runs two different courses at the same event. So for the recent SE Sprints I just combined the round 1 and round 2 results into a single CSV file and submitted. No need for any dummy event, just some minor editing of course numbers.
Had everyone in the same age class run the same course in the finals that would have been possible, but the format of the Sprint finals is that they run different courses, meaning that I can't then re-combine the age class for UKOL point purposes, without artificially adding an hour (for B finalists) or two hours (C finalists) to the times. Having done that, the results are then no good for use in ranking point calculations.
If anyone can think of a better way...
Martin Ward, SYO (Chair) & SPOOK.
I'm a 1%er. Are you?
I'm a 1%er. Are you?
-
Spookster - god
- Posts: 2267
- Joined: Fri Dec 05, 2003 1:49 pm
- Location: Sheffield
Re: British Sprints
graeme wrote:Scott wrote:I'm not convinced that "top X from each heat go into the A Final" is an inherently complex format.
I'm not sure if you're implying here that all previous organisers are indeed useless or that there is a bit of the format which isn't complex?
As far as I can tell the complexity comes from organisers trying to produce and promulgate a start list in a restricted time window on the day of the competition, instead of just publishing "7th in W50 Heat A = B Final @ 14:22" in advance. If you do that, I can't see that it make much difference whether all the heats for a single age class are run on the same course or not.
British Orienteering Director | Opinions expressed on here are entirely my own, and do not represent the views of British Orienteering.
"If only you were younger and better..."
"If only you were younger and better..."
-
Scott - god
- Posts: 2384
- Joined: Wed Jan 17, 2007 4:43 am
- Location: in the queue for the ice-cream van
Re: British Sprints
Spookster wrote:Simon wrote:These results will be for the UKOL. One guesses that someone will be adding some more results (possibly in dummy events) to get the ranking points for the heats and finals.
Exactly that. And that "someone" is me.
As soon as Scott creates the dummy events, I'll be uploading the results to them for the purposes of ranking points calculations. Meanwhile, you'll see that both the Sprints and the Middles are now showing up in the UKOL points tables (races 17 and 18 of 20 this year).
Time's running out to get the results into the ranking calculations:
Final results must be submitted electronically to the results page in the British Orienteering website within 7 days of the event. Results not uploaded within this period may be excluded from rankings calculations.
Shouldn't it actually be the UKOL results (with the merging of A/B/C finals) that is the "dummy event", rather than the ones with the courses and times people actually ran?
- roadrunner
- addict
- Posts: 1059
- Joined: Fri Sep 02, 2005 8:30 pm
Re: British Sprints
It appears that some competitors have got 3 ranking scores for the sprints!
- drobin
- light green
- Posts: 224
- Joined: Thu Dec 07, 2006 12:49 pm
- Location: Boringstoke
Re: British Sprints
drobin wrote:It appears that some competitors have got 3 ranking scores for the sprints!
In fact it seems everyone has five (!) scores: the heats and the finals twice each plus the UKOL.
Not sure whether anyone at BO now has the power to remove events from the rankings but if they do maybe they could clear the WOLF middle distance and the AIRE Fat Rascal combined prologue/sprint as neither club seems able to.
-
greywolf - addict
- Posts: 1416
- Joined: Sat Apr 07, 2007 12:45 pm
- Location: far far away
Re: British Sprints
greywolf wrote:Not sure whether anyone at BO now has the power to remove events from the rankings but if they do maybe they could clear the WOLF middle distance and the AIRE Fat Rascal combined prologue/sprint as neither club seems able to.
I reported to BO that the WOLF middle should be excluded and got this reply on the 25th Sept:
Thank you for your email. I have now set the event details as a non-ranked event.
Kindest regards
Scott Parker
I've sent Scott another email.
- Paul Frost
- addict
- Posts: 1176
- Joined: Sat Feb 26, 2005 6:25 pm
- Location: Highlands
Re: British Sprints
greywolf wrote:drobin wrote:It appears that some competitors have got 3 ranking scores for the sprints!
In fact it seems everyone has five (!) scores: the heats and the finals twice each plus the UKOL.
Three scores seems to be the answer - the email just mentions two of them twice, for some reason. The UKOL ones definitely need to be removed.
- roadrunner
- addict
- Posts: 1059
- Joined: Fri Sep 02, 2005 8:30 pm
Re: British Sprints
roadrunner wrote: The UKOL ones definitely need to be removed.
Yes.
Its surprising how, despite randomly adding an extra hour to people's times, the ranking system has still produced slightly-weird but not-totally-absurd results.
WOC2024 Edinburgh
Test races at SprintScotland (Alloa/Falkirk) and Euromeeting (near Stirling).
Test races at SprintScotland (Alloa/Falkirk) and Euromeeting (near Stirling).
-
graeme - god
- Posts: 4727
- Joined: Thu Feb 12, 2004 6:04 pm
- Location: struggling with an pɹɐɔ ʇıɯǝ
Re: British Sprints
greywolf wrote:drobin wrote:It appears that some competitors have got 3 ranking scores for the sprints!
In fact it seems everyone has five (!) scores: the heats and the finals twice each plus the UKOL.
Not sure whether anyone at BO now has the power to remove events from the rankings but if they do maybe they could clear the WOLF middle distance and the AIRE Fat Rascal combined prologue/sprint as neither club seems able to.
The problem is that it's ONLY people at BO that have the access rights to turn ranking points on or off for any event. Neither Mike Cope (National Fixtures) nor me (UKOL Coordinator) can do so. Had either of us been able to, then you wouldn't have all got 5 sets of ranking points for Saturdays races.
The correction of this is all in hand, and when the rankings re-calculation runs tonight, it should all get fixed.
graeme wrote:roadrunner wrote: The UKOL ones definitely need to be removed.
Yes.
Its surprising how, despite randomly adding an extra hour to people's times, the ranking system has still produced slightly-weird but not-totally-absurd results.
I thought the same thing (having expected that A finalists who "ran" their courses an hour faster than B finalists and 2 hours faster than C finalists might have scored 1300-1400 points).
Martin Ward, SYO (Chair) & SPOOK.
I'm a 1%er. Are you?
I'm a 1%er. Are you?
-
Spookster - god
- Posts: 2267
- Joined: Fri Dec 05, 2003 1:49 pm
- Location: Sheffield
Re: British Sprints
graeme wrote:roadrunner wrote: The UKOL ones definitely need to be removed.
Yes.
Its surprising how, despite randomly adding an extra hour to people's times, the ranking system has still produced slightly-weird but not-totally-absurd results.
True - and I suppose it says something for the robustness of the algorithm. Presumably all (or at least most) of those with 1 or 2 hours added counted as outliers, so the calculation ignored them. Then the ones in the A final got about the expected points, and those in the B and C finals extremely low scores.
- roadrunner
- addict
- Posts: 1059
- Joined: Fri Sep 02, 2005 8:30 pm
Re: British Sprints
Spookster wrote:graeme wrote:Its surprising how, despite randomly adding an extra hour to people's times, the ranking system has still produced slightly-weird but not-totally-absurd results.
I thought the same thing (having expected that A finalists who "ran" their courses an hour faster than B finalists and 2 hours faster than C finalists might have scored 1300-1400 points).
Depends which bits of the list you look at: the "bonus" points are spread amongst the A finalists so the winner's scores don't look too weird, but 1349 points for finishing 5 and a half minutes down in a sprint race is pretty silly.
roadrunner wrote:Presumably all (or at least most) of those with 1 or 2 hours added counted as outliers, so the calculation ignored them. Then the ones in the A final got about the expected points, and those in the B and C finals extremely low scores.
Probably not - the algorithm doesn't know that one hour plus is a silly score for a sprint race and just processes the numbers it's given. Adding the hours gives the distribution curve two or three humps. Where there were A, B and C races then mean time would be somewhere in the middle of the B class finishers, who would have got the mean scores, as would be anticipated. No doubt there were outliers but they would probably be at the fringes of the C race
-
greywolf - addict
- Posts: 1416
- Joined: Sat Apr 07, 2007 12:45 pm
- Location: far far away
Re: British Sprints
King Penguin wrote:... and the top end of the A ?
The course winners all scored less via the UKOL calculations than they did through the correct calculations for the finals - the difference ranges from 31 (Laura R W21) to 97 (Peter G M70) with only courses 17 and 18 having a difference > 66 points
Whereas the differences for first and last in M50C were 136 and -253
-
greywolf - addict
- Posts: 1416
- Joined: Sat Apr 07, 2007 12:45 pm
- Location: far far away
Re: British Sprints
Thinking about it further, surely this is a good test to put the algorithm through. Plainly, anyone who had an hour or more added to their time should be an outlier, so a good algorithm would discount them. You'd then be left with the 18 who actually ran the A final; some of those might also be outliers, of course, but irrespective of that the points would be the same as those actually awarded for the A final. As various people have said, that hasn't actually happened, so maybe it would be worthwhile someone going through the calculation process in detail to see why not, and whether a minor tweak would improve things. For example, would removing outliers one by one (worst first), as suggested to fix other issues, help?
- roadrunner
- addict
- Posts: 1059
- Joined: Fri Sep 02, 2005 8:30 pm
66 posts
• Page 4 of 5 • 1, 2, 3, 4, 5
Who is online
Users browsing this forum: No registered users and 12 guests