Overthe last years the topic ofpacking (influencing between runners) is perennial, but mostly, when places on the podium are involved. Forexample the pack Valstad (1.) / Berger(3.) on the WOC Long 1999 in Scotland orthe pack Khramov (1.) /Lauenstein (2.) at the WOC Long 2005 in Japan are famous. Recently the pack Tsvetkov (1.) /Efimov (10.) at the EOC 2008 in Latvia was hotly discussed forex. on World ofO, an international orienteering news site.
In this article I will show you some new perspectives on packing, developped overthe last yearand I will show you the results according to the EOC 2008.
Visualizing Packs
Following the Live Results from WOC 07 Middle Final Women I noticed (once again) that the runners were passing the intermediate controls mostly in packs. Even the final winnerSimone Niggli ran togetherwith the final third Marieanne Andersen. I wanted to get an overview on the total amount ofinfluence during a competition, that was supposed to crown the best (independently) orienteererofthe world and therefore I developed2 two perspectives/graphic visualisations on this topic. (A realtime diagramm & a Pack-Table; See this article)
My Pack-Table was very basic and consisted in a table, where so called controlpacks3 were colored. You could see, which runners influenced each otherforhow many controls. But what were the dynamics? Who was leading and who was following? This spring the norwegian orienteererHenning Spjelkavik refined and strongly enhanced this perspective.
Pack-Table by Henning Spjelkavik
The leaderofa packs ranking is understroken and italic. The position and time behind the packleaderyou get by positioning the yourmousepointeroverthe according control.
Name
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Eva Makrai
44
43
44
44
43
43
43
43
43
41
41
41
41
40
40
40
Anne Margrethe Hausken
44
26
24
21
42
42
41
40
40
39
38
36
35
35
34
34
Anna Gornicka-Antonowicz
44
32
30
27
23
33
32
30
33
33
31
30
29
28
27
27
Pam James
44
34
36
37
37
35
35
33
35
35
40
40
40
41
41
41
Tatyana Voskoboinikova
44
27
26
24
25
25
24
22
22
20
22
21
23
22
22
22
Merike Vanjuk
44
13
11
15
15
14
13
13
13
11
11
10
10
10
10
10
Line Hagman
44
37
34
39
36
39
35
32
31
31
33
31
31
31
31
31
Iliana Shandurkova
44
25
22
22
21
27
33
38
38
38
37
37
37
37
38
38
Zsuzsa Fey
44
31
29
31
40
41
42
42
41
43
42
42
42
42
42
42
Helen Winskill
44
41
40
38
35
38
36
34
33
34
34
34
34
34
35
35
Ildiko Szerencsi
44
35
33
35
34
32
40
41
42
43
43
43
43
43
43
43
Minna Kauppi
44
3
2
2
2
8
6
5
5
4
4
4
4
4
4
4
Christiane Trobe
44
19
41
42
41
40
39
39
39
40
39
39
39
39
39
39
Zanda Abzalone
44
28
27
30
27
22
30
27
29
28
28
33
33
33
33
33
Seline Stalder
44
14
14
9
9
11
11
31
30
30
30
28
28
27
26
26
Rachael Elder
44
33
31
29
26
19
19
23
24
24
23
22
22
23
23
24
Zdenka Stara
44
8
8
7
13
7
7
9
8
8
8
11
11
11
11
11
Jo Allison
44
44
44
43
44
44
44
44
44
44
44
44
44
44
44
44
Tania Robinson
44
40
39
40
38
37
38
36
37
37
35
35
36
36
36
36
Grace Elson
44
23
21
28
29
28
29
26
26
25
27
32
32
32
32
32
Indre Valaite
44
29
23
18
19
16
22
19
16
15
18
17
17
17
17
17
Martina Rakayova
44
22
20
26
28
30
27
25
27
29
29
25
25
21
21
21
EstherGil Brotons
44
21
19
25
23
34
37
37
36
37
36
38
38
38
37
37
Eva Jurenikova
44
11
12
12
11
18
17
12
14
12
12
12
12
12
12
12
Liis Johanson
44
36
36
36
32
29
26
24
23
26
26
27
27
29
29
29
Celine Dodin
44
18
17
20
18
17
14
18
15
13
13
15
15
15
14
14
Mariya Spasyuk
44
16
18
17
14
21
20
17
17
19
19
18
18
26
25
25
Radka Brozkova
44
12
13
13
10
10
9
10
10
9
9
8
8
8
8
8
Jenny Johnson
44
24
25
23
21
27
25
21
20
21
21
24
25
24
24
24
Capucine Vercelotti
44
8
6
14
12
15
12
11
11
16
17
26
26
25
28
28
Martina Fritschy
44
38
38
35
33
31
28
29
28
27
25
21
20
19
21
20
Signe Soes
44
10
9
16
25
20
18
14
18
17
15
14
14
15
15
15
Aija Skrastina
44
39
37
33
31
24
23
21
21
22
20
19
19
18
18
19
Natalya Korzhova
44
30
28
19
16
12
16
17
19
14
14
16
16
16
16
16
Emma Engstrand
44
3
5
4
4
4
5
6
6
7
7
7
7
7
7
7
Sandra Pauzaite
44
42
42
41
39
36
31
28
25
23
24
23
21
20
19
19
Annika Billstam
44
15
15
10
7
6
15
35
34
32
32
29
30
30
30
30
Marianne Andersen
44
4
3
5
5
3
3
3
3
3
3
3
3
3
3
3
Simone Niggli
44
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Hanny Allston
44
20
32
32
30
23
21
15
12
18
16
13
13
13
13
13
Paula Haapakoski
44
6
8
6
17
13
10
8
9
10
10
9
9
9
9
9
Tatyana Riabkina
44
17
16
11
8
9
8
7
7
6
6
5
5
6
6
6
Heli Jukkola
44
5
4
3
3
2
2
2
2
2
2
2
2
2
2
2
Helena Jansson
44
9
10
8
6
5
5
4
4
5
5
6
6
5
5
5
What a development! You now can see who ran with whom, who was leading and somehow even the effect on the runners ranking. You get a really good impression ofthe pack dynamics ofa competition. But is this impression true? Do only followers profit ordo the leaderprofit too? To which amount? Is the role ofleaderreally assignable? And anyway, how long where they running together(one control can take as much time as 6 others)...
Packfigures™
To furtherrefine the analysis I programmed a .php-Appliction I call Packfigures™. With Packfigures™ I try to bring the general impression ofthe Pack-Table down to figures. Forthat I compare every runners speed in- and outside packs to calculate, what I call the boost, the effect ofthe pack and the time gained by it. Fordetails check out the footnote.4
Packfigures ™ forWOC 2007 Middle Final Women
-> A proper descriptions of the columns content you get by puting the
mouse over the columns headers.
-> A runners leg- and controlpackmates you get by placing the mouse over the controls
number
As you can see packs did not severly influence the top results of last years WOC Middle Final Women. Niggli and Jukkola did not perform better in packs. Andersen profited from Nigglis speed at the end. Kauppis boost consists in a single control where she ran a good pace before really spoiling her lone-speed-figure at the fifth control. Jansson did a good solo performance. Riabkina really took a profit from the pack with Jukkola and Alston always in front. Where as Engstrand could not keep the speed up in the second part of the race, where she had to run alone. And so on. There are some athletes performing better in packs, some being hardly influenced and some even disturbed by packs. For example Celine Dodin mentioned, she really was happy to get rid of the pack for she felt performing worse than usual (calculated boost: -5.12%). Here it's to add that Packfigures™ is most of all usefull for the analysis of longdistance competitions, where the task for keeping the speed and the concentration high is heavier.
As an example I analyzed the EOC 2008 Long Final Men, a competitions mentioned in the introduction, where Tsvetkov and Efimov build a pack for more than half of the race. By the way this competition was the first visualized by Hennings Packtable and it was bullseye :-). Now, Packfigures™ shows that Tsvetkovs boost was only about 2.4% where Efimovs was 14% (and Nikolovs was 16%). His packless runtime still would be enough for winning (about the same packless runtime as Wingsted). Wingsteds boost in pack with Lucan was even smaller (1.2%). Both lead packs I call H-Packs for hierarchical packs, where the packs benefit is uneven among the runners. The roles of the little profiting leader and the profiting follower can be assigned.
The most severe pack according to the final result on EOC 2008 Long Final Men was the one with Hubmann and Johansson, where two top runners build a pack, both pushing and trying to take the lead for more than fourty minutes. They thus formed a E-Pack with significant benefits for both (Equal-Pack; Hubmann 5.31%, Johansson 9.67%).
More Analysis
Here I link all the analysis made with Packfigure™
Note: Feel free to conact me if you could provide me with more intressting data
Conclusions
These analysis show, that packs can have a variety of effects on the influenced runners. Some might profit a lot, some a little, some none and some even start to lose their concentration (Bertuks/EOC 08, Dodin/WOC07). Fact is that packs generally boost a runners performance. The boosts size depends on the runners solo performance as on the hierchical position of a runner inside the pack (leader, co-leader or follower). Generally it is the follower, who gets the biggest benefit, than the co-leader and last but not least the leader. As at championchips we don't see any follower in the top 6 but leaders and co-leaders, the main focus should be on avoiding most of all E-Packs and then packs in general.
This goal is severly contradicted by the organisational framework for international championchips defined by the IOF: runner start in short 2 minutes intervalls and the start in reverse order according to their qualification result. This means that due to the start intervall is very likely they catch up each other und due to the start order it is very likely they stay together and form an E-Pack. Further it is obvious that the start order ruins the butterflies effectivity. Therefore the IOF really should revise/reflect the competitions framework, before focussing on separation methods really changeing the nature of orienteering.
2 Not knowing that one already existed in Splitbrowser
3 Controlpack: Two runners punching the same control not less than 15 seconds apart
4Some Details: I make a distiction between controlpacks and legpacks. The first is fullfilled, when two runner punch at the same control not more than 15 seconds apart. I assume these 15 seconds as the margin, where a runner gets some usable information from his frontrunner attacking the control5. A legpack is achieved when the same runners are in the same controlpack for two sequent controls. For the runners speed outside packs I take the runners splittimes of all legs being neither influenced by control- nor legpacks and set them into relation with the best split of all runners outside any pack. For the runners speed in packs I put all the runners split compiled in legpacks in relation to the according top splits of all runners outside any pack. Now the boost is the relation of these two speeds. The gain is time gained due to the boost during the runners time in the legpack. These figures can be unprecise, if a runners performance on the legs not taken into account (due to a controlpack) differs significantly from the rest of his performance. Ignoring the controlpacked legs also leeds to a conservative value for the gain: potential benefits from controlpacks are ignored. Out of the gain I calculate the packfree runtime and the packfree ranking for all runners. These figures are not entirely fantastic and indeed have a provocative nature. Discussion: By relating the runners speed with the top solo splits the model takes into account effects like holding up the speed on the last part of a long distance race. Instead the model is not capable to handle any mistakes shortened or not done due to packs, where made mistakes are taken into account.6 So there is a certain oddity.
5 Henning Spjelkavik showed in an other study, that you can find an agglomeration effect of preceding runners in this intervall.
6 An example: on EOC 08 Long Final Men. Matthias Merz makes a mistake at the 23rd control where he lost 56 seconds to the top split. This mistake is fully taken into account to his solo performance. Without this mistake his performance would have been good enough to win. Instead all his direct concurrence attacked the same control in company. On the other hand Tsvetkov always leading his pack unexpectly punches the 21st control at the very end of his pack and loses 22 seconds on this leg. Here only this 22 seconds are taken into account and it seems that somehow his pack saved him.
