Timing

General slalom chatter...rant about the bad, rave about the good
oldandslow
Posts: 180
Joined: Sun Dec 18, 2005 9:11 pm
Location: Peak District

Timing

Post by oldandslow » Tue Sep 11, 2018 10:08 pm

Timing
At Division 2 to 4 events that are not using an electronic timing beam, then timing should be calculated to the nearest second (not 0.1 second). The reason being that a change of start or finish judges or shift in their position could easily effect timing by more than 0.5 seconds. This would obviously result in more ties and therefore the final position will be compared to the non-scoring run rather than tenths of second timing.


Selection of Results from a div 2 event in 2018

Comp Pos Run 1 Run2 Best Points
1 18 118.4 117.1 117.1 345
2 19 171.9 117.2 117.2 336
3 20 117.3 119.4 117.3 327
4 21 117.3 171.9 117.3 318
5 22 168.2 117.5 117.5 309
6 23 118.0 117.8 117.8 300
7 24 120.8 117.8 117.8 291
8 25 120.9 117.9 117.9 282


How the results change with a 1 second accuracy…
Comp Pos Run 1 Run2 sec Points
1 18 118 117 118 345
6 18 118 117 118 345
3 20 117 119 119 327
7 21 120 117 120 318
8 21 120 117 120 318
5 23 168 117 168 300
4 24 117 171 171 291
2 24 171 117 171 291

Therefore a nod of the head or the wriggle of a bottom causing a 0.1 second timing error, a 5 place change of position and a 45 point discrepancy!

Proposal by Donna Hawkins of Matlock Canoe Club

C33.4.2 Timing
Division 2 to 4 events using any other start and/or finish for the whole event should be timed to 1 second accuracy.
Life is what happens when you're making other plans.

JimW
Posts: 570
Joined: Thu Oct 08, 2015 2:17 pm
Location: Pinkston

Re: Timing

Post by JimW » Wed Sep 12, 2018 2:15 pm

I understand that the WWR committee spent a bit of effort studying this (or at least, one mathmatician amongst them did) and they do use different accuracies to slalom although the reasons are not quite as you have outlined. You might want to reference GB WWR rule 17.6 and see if you can research the background of it from the current committee.

Start and finish judges shifting position should not affect the timing at all, they should be sighting through 2 points (poles normally) and pressing the button when the paddlers body breaks the line through the points. It does not matter if the judge moves their body position, they will have to move their head back in line as the paddler approaches because there is no other way they can get the 2 poles in line. It is best if the near pole is not too close to the judge so as it appears thick enough that the far pole disappears behind it as that could allow for a slight difference in finish line angle.

The reasons for different accuracy when using a stopwatch I beleive are to do with reaction times, not the judges moving around.

For many years the speed of ships has been determined using a measured mile course which is essentially a start and finish line set 1 (or 2) nautical miles apart using huge white posts (or lit towers) on the shore with the ship normally about 4 cables from the shore. The recorder on the bridge starts a stopwatch when the first set of posts come into alignment and stop it when the second set reach alignment, the test is repeated 3 or 5 times (depending on the contract) with runs in both directions to negate the affects of wind or tide. Having conducted a few of these over the Skelmorlie mile when I was in the shipyards, I can tell you that the times for runs in the same direction are always within a few tenths and that is with a considerable variation possible in the distance from shore and exact heading of the ship.
The 2 post alignment method is sound! Reactions of judges are questionable :)

Dee
Posts: 1444
Joined: Sun May 08, 2005 8:34 pm

Re: Timing

Post by Dee » Wed Sep 12, 2018 4:22 pm

This would inevitably mean a lot more cases of second runs coming into play which I feel may be unfair in the lower divisions. I think it is relatively rare when timing is to 100th.

Maybe we should be looking at ditching the use of second runs to distinguish and just say that in the case of a tie, both paddlers get the higher points or, maybe share the points for the two positions. With your nearest second example this could result in quite a lot of paddlers getting the same points!

I think it is unusual to get quite so many paddlers so close together at lower division level though
Kit Washer, Entry Clerk, Chauffeur, Reluctant Organiser, Online Entry Advocate .....
Anything I post under this user is my personal opinion; I am not posting as a member of the Slalom Committee!

oldandslow
Posts: 180
Joined: Sun Dec 18, 2005 9:11 pm
Location: Peak District

Re: Timing

Post by oldandslow » Wed Sep 12, 2018 4:28 pm

Dee, This closeness in timing is extremely common at Matlock amongst the div 2 k1M. Have a look - it happens for almost every second at the top end.
Life is what happens when you're making other plans.

Dee
Posts: 1444
Joined: Sun May 08, 2005 8:34 pm

Re: Timing

Post by Dee » Wed Sep 12, 2018 4:43 pm

Is it just Matlock though, or is it a widespread issue.
Not saying we should ignore the issue just the approach might depend on how widespread it is.

My point still stands that, truncating the times to seconds will result in a lot of second runs being used for placing which I'm not in favour of.

Do Matlock not use electronic timing?
Kit Washer, Entry Clerk, Chauffeur, Reluctant Organiser, Online Entry Advocate .....
Anything I post under this user is my personal opinion; I am not posting as a member of the Slalom Committee!

oldandslow
Posts: 180
Joined: Sun Dec 18, 2005 9:11 pm
Location: Peak District

Re: Timing

Post by oldandslow » Wed Sep 12, 2018 10:13 pm

No, Matlock use the tutti.

I don't know if it's a common problem. I'm sure the div 2 K1M will know though.

I still think it would be fairer to use second runs than lose several places because of a fraction of a second. two consistent runs would indicate a better paddler in my mind

Perhaps it could be a half second accuracy?
Life is what happens when you're making other plans.

JimW
Posts: 570
Joined: Thu Oct 08, 2015 2:17 pm
Location: Pinkston

Re: Timing

Post by JimW » Thu Sep 13, 2018 1:21 am

oldandslow wrote:
Wed Sep 12, 2018 10:13 pm
No, Matlock use the tutti.

I don't know if it's a common problem. I'm sure the div 2 K1M will know though.

I still think it would be fairer to use second runs than lose several places because of a fraction of a second. two consistent runs would indicate a better paddler in my mind

Perhaps it could be a half second accuracy?
I would need to understand exactly why WWR chose the accuracy they did before I could comment on whether it is right or wrong to change the accuracy used for slalom, but as a paddler I would prefer more accuracy so that the places are based on the fastest run time.

I have never seen anyone question if the timing we use is suitable for the accuracy of the results, let along complain that they might have lost places/points because of it.

I have often seen very good paddlers make a mistake, or get a judging decision go against them on one of their runs - which is why we have 2. If we get to a situation where you always need two consistent runs it will lead to even more protests because paddlers won't be as prepared to let a bad run go. It will also stop paddlers with a good first run trying something different to see if they can go even faster on the second run at risk of missing or touching gates that they didn't before.

Nick Penfold
Posts: 338
Joined: Thu May 27, 2004 8:21 pm

Re: Timing

Post by Nick Penfold » Thu Sep 13, 2018 2:55 pm

We used to time Div 2 to Div to whole seconds as standard. The change to 0.1 second came in 2009, and the rule is now a little strange (the jury has discretion, but no alternative).
Up to 2008:
33.4.2 Division 2 to 4 events using any other start and/or finish* for the whole event shall be timed to 1 second.
From 2009:
33.4.2 Division 2 to 4 events using any other start and/or finish* for the whole event shall be timed to 0.1 second with the approval of the jury.
The jury's alternative if they don't "approve" isn't stated.
The change came in via motion 6.13 at the 2008 ACM. The reasoning isn't recorded in the minutes, but to the best of my recollection we were getting ridiculous numbers of paddlers on equal points, and resulting points inflation, even after taking the other run into account. Oldnslow's example above shows three equal-best-scores among just eight results, and I can remember seeing four or five paddlers on the same points:
Comp Pos Run 1 Run2 sec Points
1 18 118 117 118 345
6 18 118 117 118 345
3 20 117 119 119 327
7 21 120 117 120 318
8 21 120 117 120 318
5 23 168 117 168 300
4 24 117 171 171 291
2 24 171 117 171 291

* than electronic

djberriman
Posts: 806
Joined: Tue Apr 03, 2007 8:19 pm

Re: Timing

Post by djberriman » Thu Sep 13, 2018 11:07 pm

I do tend to agree with this, its been incredibly close at the top of Div 2 this year, factor in low enties (<30 in K1M) and the only ones getting worthwhile points are 1st and 2nd. I'm dubious that the timing is accurate especially given different start/finisher during the day and thus its probably luck of the draw whether you end up in the top 2. If the starter/finisher changes during a class who is to say they are timing exactly the same and have the same reactions. If it happens just before say the last 5 down it could have a dramatic effect.

I guess there should be a stipulation that starter/finisher only changes at between classes but then would that work as paddlers at these events run out of order/class quite a lot.

I've thought for a while that possibly all events should be scored out of 45 (or some such number) if there are less than 45 paddlers just to reward those who do turn up and make it worth the expense etc. In this way the top 3 would always get worthwhile points (or possibly more depending on where the limit is set). I remember going to Tully and Div 2 K1M had just 13 paddlers, we pretty much knew the order we would finish in, only 1st got anything, the rest may have well have saved the 12 hours driving and expense.

Would be nice if we could have beams (or something else) at more events to produde better timing.

I guess the problem has got worse since courses were shortened from 120 secs to 90 secs, statistically the run times are probably closer.

Steve Agar
Posts: 56
Joined: Wed Jun 17, 2015 9:30 am

Re: Timing

Post by Steve Agar » Fri Sep 14, 2018 4:57 pm

A bit of simple maths - always subject to assumptions and open to argument!

A slalom course should average 300m distance through the gates and take about 100 secs for K1M.

Average paddler speed over the course is therefore about 3m/second.

A time difference of 1 second between competitors therefore equates to a difference between finishers of about 3m, or nearly a boat length.

A time difference of 0.1 secs equates to a distance difference of about 30cm or 1 foot. Pretty clear to say who should win.

A time difference of 0.01 secs equates to a distance difference of 3cm or about 1 inch.

Recording a time to one tenth accuracy is very difficult without measuring to hundredths and rounding!

Given the vagaries of how water changes down a course between competitors, I find it difficult to believe that races should be judged to be won or lost on the basis of 1/100 second, but apparently we have the technology that allows us to do this. Should two paddlers approaching a finish line separated by less than 3 cm really be differentiated? However, at one tenth of a second resolution, our human starter and finisher would have to be more than 30cm out in the paddlers start and finish position to affect the recorded time. (Standing starts are a different matter!) If we use the two-pole method, and we make sure the poles are actually on the marked finish line (which when using trees on the bank as markers isn't always the case!) I would have thought that timing to tenths (after rounding) therefore provides a reasonable discrimination in deciding who wins. Without checking, I'd be surprised if many div 3 events are decided by as little as one tenth of a second so the argument loses some relevance anyway but I'm happy to be corrected. Given the current standard at div 2, I think trying to separate paddlers by whole seconds rather than tenths would cause more problems than it solves. What we know is that promotion depends on consistency across several events, and losing out on points by 1/10 of a second this week is likely to be compensated by gaining by 1/10 at the next event. Over a season, I'd be surprised if anyone lost out on promotion because their place was determined by timing to tenths rather than whole seconds. At Div 1 and Prem, every 1/10 matters (although I maintain my reservations about the relevance of hundredths given the other variables at play in our sport where water plays an essential role and wind gets itself involved!).

I do wonder why we worry about timing accuracy so much when penalties rely so much on human judgement, and will remain to do so regardless of whether we use laser technology, GPS tracking or atomic clocks for timing. I know far more occasions when competitors have lost out (or gained) by poor, or more usually, inconsistent judging than the odd tenth of a second difference in recorded time. But that's another matter :-)
Last edited by Steve Agar on Sat Sep 15, 2018 11:28 am, edited 1 time in total.

djberriman
Posts: 806
Joined: Tue Apr 03, 2007 8:19 pm

Re: Timing

Post by djberriman » Sat Sep 15, 2018 10:02 am

Problem is as I mentioned manual timing relies on all the people who do start/finish timing in the same way with the same accuracy and doing so all day.

Also it means start/finish should sit in the same place and be under the start/finish line. I've seen start sat about a metre up from a start line. The course design meant you could go through start anywhere within 10m of the bank. Obviously there was a sweet spot but depending on where you crossed the start line it would therefore lead to quite an inaccuracy. I don't remember seeing any poles on the line at that event either so presumably they were just using an object on the far bank to decide when you crossed.

You are probably right it evens out over a year but non the less frustrating.

Perhaps need something in the organisers pack about rolling starts and manual timing, start/finish lines/poles etc. It could also include a summary of the rules regarding number of gates/type and short courses etc.

Nick Penfold
Posts: 338
Joined: Thu May 27, 2004 8:21 pm

Re: Timing

Post by Nick Penfold » Sun Sep 16, 2018 5:41 pm

I've seen start sat about a metre up from a start line.
The solution to this is putting the judge in the right position, not adjusting the timing standards!

JimW
Posts: 570
Joined: Thu Oct 08, 2015 2:17 pm
Location: Pinkston

Re: Timing

Post by JimW » Mon Sep 17, 2018 10:22 am

Nick Penfold wrote:
Sun Sep 16, 2018 5:41 pm
I've seen start sat about a metre up from a start line.
The solution to this is putting the judge in the right position, not adjusting the timing standards!
Agreed.
The people who know/understand this may not be around at every race, perhaps we need to make sure we have something formal in the course design notes and instructions for start/finishers to ensure that a suitable line of sight is established in the first place and that the volunteers know how to use it?

Reminder sheets to print with the start/finish sheets and put on the boards?

alan1nckc
Posts: 45
Joined: Thu Jun 28, 2012 6:19 pm

Re: Timing

Post by alan1nckc » Thu Nov 15, 2018 12:21 pm

<quote> A slalom course should average 300m distance through the gates and take about 100 secs for K1M.</quote>

On the Nene whitewater course, our 18 gates cover about 250 metres and take around 110-120 seconds for K1M.

Should we shorten the course to get to 100 seconds, or lengthen it to get to 300M? Or make all the gates downstream so it's a sprint?

K1M results show 9 equal places out of 34 on 1-second timing.
<pre>
Pos Age --First Run-- --Second Run-- Best Points
1 J12 119.7 4 123.7 114.4 0 114.4 114.4 500
2 J14 118.7 2 120.7 114.8 2 116.8 116.8 485
3 J14 117.9 2 119.9 117.1 0 117.1 117.1 471
4 J18 119.2 0 119.2 119.1 0 119.1 119.1 456
5 J14 124.7 0 124.7 120.8 0 120.8 120.8 441
6 J16 122.3 0 122.3 130.7 0 130.7 122.3 426
7 J16 122.5 0 122.5 122.5 2 124.5 122.5 412
8 M 123.8 2 125.8 123.3 2 125.3 125.3 397
9 J12 125.0 6 131.0 126.5 0 126.5 126.5 382
10 M 126.7 0 126.7 126.6 0 126.6 126.6 368
11 J14 122.3 6 128.3 123.8 4 127.8 127.8 353
12 J14 126.5 2 128.5 128.7 0 128.7 128.5 338
13 J14 124.1 6 130.1 128.6 10 138.6 130.1 324
14 J14 128.9 2 130.9 130.8 4 134.8 130.9 309
15 J12 130.0 2 132.0 127.6 8 135.6 132.0 294
16 M 134.1 2 136.1 132.7 0 132.7 132.7 279
17 M45 139.1 6 145.1 133.0 0 133.0 133.0 265
18 J16 128.0 6 134.0 127.3 6 133.3 133.3 250
19 J12 133.8 2 135.8 134.0 0 134.0 134.0 235
20 M35 141.9 2 143.9 132.9 2 134.9 134.9 221
21 J16 133.8 2 135.8 143.6 2 145.6 135.8 206
22 M60 139.2 0 139.2 136.0 2 138.0 138.0 191
23 M 144.2 0 144.2 137.0 2 139.0 139.0 176
24 M 136.1 4 140.1 131.9 8 139.9 139.9 162
25 J12 138.0 2 140.0 152.4 50 202.4 140.0 147
26 U23 138.7 2 140.7 138.8 2 140.8 140.7 132
27 J12 148.0 8 156.0 131.5 10 141.5 141.5 118
28 J12 136.3 6 142.3 136.1 10 146.1 142.3 103
29 J12 141.1 6 147.1 140.0 4 144.0 144.0 88
30 J12 142.4 2 144.4 152.5 8 160.5 144.4 74
31 M 148.4 4 152.4 145.5 2 147.5 147.5 59
32 M 146.6 8 154.6 144.6 8 152.6 152.6 44
33 M 155.9 6 161.9 151.8 4 155.8 155.8 29
34 J14 154.3 52 206.3 153.4 6 159.4 159.4 15
</pre>

Nick Penfold
Posts: 338
Joined: Thu May 27, 2004 8:21 pm

Re: Timing

Post by Nick Penfold » Thu Nov 15, 2018 6:20 pm

I don't know but as Steve estimated:
A time difference of 1 second between competitors therefore equates to a difference between finishers of about 3m, or nearly a boat length.
A time difference of 0.1 secs equates to a distance difference of about 30cm or 1 foot. Pretty clear to say who should win.
A time difference of 0.01 secs equates to a distance difference of 3cm or about 1 inch.
Steve's estimate is conservative, because it's based on the average speed including breakouts and most paddlers will cross the finish line going quite a lot faster than their average speed - unless they're drifting. I think Steve's distances travelled will still be near enough true, even with Alan's times and distances.
Using a button press you can't consistently time to .01 second, and we don't need to anyway.
You can time to 0.1 second
Timing to 1 second - a whole boat length - doesn't make sense to me.

Post Reply