info@orienteering.ie

Forum

Notifications
Clear all

League Point Calculations & Outliers

46 Posts
7 Users
0 Reactions
10.7 K Views
Posts: 262
(@gradient)
Reputable Member
Joined: 11 years ago

Stuart said:

Are there specific examples of people that have been \'hard done by\' by the current system?

Yes. In the league just finished, Eoghan Whelan won the men\'s category on Blue with three firsts, 1 second and two fourth places, while Brendan McCarthy came second with 5 firsts and a second!

Stuart said:

Considering that only a few of us are having this discussion, is there really any appetite to change the system?

Actually, I know that people have been followng at least some of this thread from comments made to me at the event last night. The fact that only a few of us are having this conversation really doesn\'t tell us anything about what people\'s opinions on the matter are. You are using a fallacious argument called an appeal to silence.

Reply
Posts: 262
(@gradient)
Reputable Member
Joined: 11 years ago

Ok, I just noticed that there was a correction to the points awarded for the Curragh after the posting I was looking at, and Brendan did in fact win, but it doesn\'t really change the issues. With the change in one runners time Brendan apparently moved 32 points ahead of Eoghan.

My first comment is that 30 points doesn\'t seem to me to fairly reflect Brendan\'s better performance of 5 wins and 1 second against three wins, 1 second and two fourths. Especially given the overall points scores at over 7600 (roughly only a 0.4 percent difference).

I don\'t know where the runner whose time was changed was positioned originally but the current results are showing them at the bottom of the results. If they were originally showing as ahead of Brendan then they would have been first and Brendan would have been second and in that case Brendan would have lost to Eoghan as per the original final league places given.

In that case Eoghan\'s result remains as before, i.e. 1, 1, 1, 2, 4, 4 (he didn\'t run at the Curragh) and Brendan\'s is 1, 1, 1, 1, 2, 2 which I think by any estimation is still much better, but the points system puts him second! The points system completely fails to accurately represent Brendan\'s superior performance.

Reply
Posts: 262
(@gradient)
Reputable Member
Joined: 11 years ago

Another example is the orange course in the league just gone. First in the points table is Niamh Browne with 7672 points, closely followed by Fionnuala Rowe on 7620. But Fionnuala won her course six times whereas Niamh\'s best 6 results are 1 win, 4 seconds and a third. Both fantastic results by any standard but should Niamh be placed ahead of Fionnuala?

Reply
Posts: 100
(@rshort)
Estimable Member
Joined: 11 years ago

@psmythirl the original position of the changed competitor was also last place, but with a much much larger margin (they never cleared and were one of the last starters).

It\'s a bit of a contrived scenario, but I created one that I think indicates why some people would be pro points as opposed to position.

Who should win in this scenario? Say 4/5 to count. Runner A clearly has more wins but Runner B has one very decisive one. You can probably argue both ways.

Event 1                         Event 2                          Event 3                        Event 4                          Event 5
Runner A 10:00             Runner A 10:00              Runner A 10:00             Runner A 10:00              Runner B 8:00
Runner B 10:01             Runner B 10:01              Runner Z 10:01             Runner B 10:01              Runner A 10:00
Runner C 10:02             Runner C 10:02              Runner C 10:02                                                  Runner C 10:05

Personally I feel that in orienteering, when we run individually, beating somebody by a small margin can often be more down to luck as much as anything else. If you race in a mass start it\'s a different scenario and I would argue for a different point system.

Reply
Posts: 74
 Rory
(@roryajax)
Trusted Member
Joined: 7 years ago

I think you\'ve made your case for an actual comparison of some different systems for a league\'s results.
I\'ll suggest 2 that can surely be done simply in Excel or in PHP, and which represents probably both extremes of the argument:
- Median Absolute Deviation : take the median time per event, calculate the median (absolute) differences of all finishers time from the median, rescale by 1.48; substitute this measure for the standard deviation, and the median for the mean
- The \"Formula 1\" system with 1000 points for the Winner decreasing by 10 points per place (the purest rank-based system there is)
I\'ll work on these some evening this week hopefully.

Reply
Page 7 / 10
Share:

Orienteering in Ireland
Orienteering Ireland, Irish Sport HQ, Blanchardstown
D15 DY62, Ireland