The CKO Rankings were first conceived in 2000 when it became apparent that the sanctioning organizations were not adjusting their rankings based on results. The largest promotional company had created a total eclipse of the sport and their tournament champion was generally recognized as world champion. The CKO set out to fill in the gaps left by this giant promotional company and the sanctioning organizations.
One of the interesting aspects of tracking kickboxing activity with an objective list is that every reader will compare it to his opinion. I’ve read many critiques. Some of these show a lack of understanding for the CKO ranking procedures. The most common example of this is, of course, when forum members assume that the K-1 (the previously mentioned promotional company) Grand Prix champion will automatically take the #1 rank. Nothing in the CKO procedures suggests such a privilege to any promoter.
Other members disagree with how the rankings are formed. In these cases, it is not so much a forum member’s alternative being compared to the CKO ranking procedures. These members never have an alternative system in mind. They simply see RED ranked lower than BLUE and disagree with the list and conclude that the system is flawed. They never have an objective system in mind to ensure that RED would be higher than BLUE.
Diplomacy has become a tool of choice to take issue with the results of the CKO system. In short, they feel that if they can organize enough members to disagree with the CKO results that they can create the illusion of everyone outside the CKO being in agreement. This is not the case of course. For true agreement to take place, unifying to disagree is far from enough. They will need to all agree on one list. The more people they organize, the more compromise they will have amongst themselves. Compromise is not agreement. Not quite.
To rebut the diplomacy attacks, the CKO has always made it clear that fans and fighters are two different groups. While fans have complained that BLUE was ranked too high, fighters have only had good things to say once they realized that the list was objective. That said, to counter the claims that fans disagree with the CKO rankings, I set out to measure statistically if there was any truth to this. To avoid bias, I enlisted the help of a third party to select between 5 and 10 kickboxing pundits. I would ask for 10 and hope for 5 to respond. There are many fans of the sport, but the pundits know the sport better than most. They know the fighters, the coaches, and the promoters. They see the K-1 as a part of the sport, not the sport itself. They are experts.
The 3rd party consultant I chose was (oYo), or Boobs as many call him. His online pseudonym is designed to look like a pair of breasts, something men have always enjoyed since birth. Boobs has been a forum moderator on several message boards and is usually one of the first to report any developments in the sport. He is based in Japan and therefore has a frontline perspective of the game. He now is a partner in www.kakutougi.info and has attracted members to his forums due to the relationships he has built over the years. Other forums may boast larger memberships, but www.kakutougi.info enjoys a good class of pundits. Having Boobs choose the 10 pundits was important. I didn’t want my own preferences to cloud my judgment and produce a rosy correlation.
Reliability vs. Validity
Two terms that everyone interested
in rankings systems will have to know are reliability and validity. Reliability
refers to the integrity of the system. If the same action happens twice
and produces different results, you have a flaw in reliability. Mark Hunt
and Frans Botha both defeated Jerome LeBanner. Because the CKO aims for
reliability, the result was the same. They each took LeBanner’s rank. Validity,
however, is a measure of how meaningful the results are. The K-1, for
instance, holds
tryouts in Las Vegas
and the participants run through a series of tests. If the test results
match very closely to the subsequent match ups, then the tests can be
called valid. To determine if the CKO is a valid list we must compare it
to what it claims to measure. The CKO has overtly denied any suggestion
that the rankings are a best to worst list. If anything, the list reflects
the success of the fighters. That therefore became the question I put to
the pundits. Giving the CKO Top 20 list, I asked to put the fighters in
order based on success, activity and quality of opposition. A few returned
with questions regarding the value of last year’s success in comparison
to the previous years. The answer was, “You
decide.”
Spearman’s rho is a statistical
calculation that takes two rankings and produces a numerical relation from 1
to -1. A score of 1 means that the lists are identical (1,2,3,4 vs.
1,2,3,4). A
score of -1 means that the lists are reversed (1,2,3,4 vs. 4,3,2,1). A score of
0 (zero) means that there is no relation whatsoever between the two lists.
|
Range |
Conclusion |
|
0.900 to 0.999 |
Very Strong Agreement |
|
0.800 to 0.899 |
Strong Agreement |
|
0.700 to 0.799 |
Agreement |
|
0.600 to 0.699 |
Moderate Agreement |
|
0.500 to 0.599 |
Moderate Disagreement |
|
0.400 to 0.499 |
Disagreement |
|
0.300 to 0.399 |
Strong Disagreement |
Here are the most recent correlations:
2007
= 0.699
2006 = 0.785
2007
Below
are the 2007 rankings produced by the pundits (PA to PF).
A1
is the real average of PA to PF. It reflects how close fighters are to
each other by ignoring the rungs of the ladder. Where a normal ranking
would be 1-2-3-4, the chart below will show 1- 2-4-4.5
A2
is the average of PA to PI adjusted to form a ranking where each rank is
given the appearance of being equally spaced. Note that ranks 1 9 and 20
were tied and thus, for statistical purposes, the rank 1 9.5 was given to
each.
|
Year
2007 |
CKO |
PA |
PB |
PC |
PD |
PE |
PF |
A1 |
A2 |
|
Sem
Schilt |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
Remy
Bonjaski |
2 |
4 |
3 |
6 |
4 |
3 |
4 |
4 |
3 |
|
Hong
Man Choi |
3 |
8 |
7 |
9 |
8 |
7 |
13 |
8.66 |
8 |
|
Glaube
Feitosa |
4 |
6 |
4 |
3 |
3 |
5 |
6 |
4.5 |
4 |
|
Ray
Sefo |
5 |
7 |
7 |
4 |
6 |
8 |
5 |
6.17 |
7 |
|
Jerome
LeBanner |
6 |
3 |
8 |
5 |
7 |
6 |
3 |
5.33 |
5 |
|
Stefan
Leko |
7 |
5 |
6 |
7 |
5 |
4 |
8 |
5.83 |
6 |
|
Peter
Aerts |
8 |
2 |
2 |
2 |
2 |
2 |
2 |
2 |
2 |
|
Bjorn
Bregy |
9 |
11 |
12 |
10 |
15 |
12 |
16 |
12.66 |
12 |
|
Siala-Mou
Siliga |
10 |
17 |
11 |
17 |
18 |
20 |
12 |
15.83 |
15 |
|
Dzevad
Poturak |
11 |
14 |
19 |
18 |
19 |
16 |
19 |
17.5 |
19.5 |
|
Chalid
Arrab |
12 |
9 |
9 |
8 |
9 |
9 |
9 |
8.83 |
9 |
|
Doug
Viney |
13 |
18 |
20 |
14 |
12 |
14 |
20 |
16.33 |
17 |
|
Tsuyoshi
Nakasako |
14 |
19 |
17 |
19 |
16 |
18 |
18 |
17.33 |
18 |
|
Alexander
Ustinov |
15 |
10 |
14 |
12 |
11 |
13 |
17 |
12.83 |
13 |
|
Jurgen
Kruth |
16 |
15 |
13 |
16 |
20 |
17 |
15 |
16 |
16 |
|
Kaoklai
Kaennorsing |
17 |
20 |
18 |
20 |
17 |
19 |
11 |
17.5 |
19.5 |
|
Akio
Mori |
18 |
12 |
15 |
15 |
10 |
11 |
10 |
12.16 |
11 |
|
Peter
Graham |
19 |
13 |
16 |
11 |
13 |
15 |
14 |
13.67 |
14 |
|
Badr
Hari |
20 |
16 |
10 |
13 |
14 |
10 |
7 |
11.67 |
10 |
Below
is the chart describing the correlation between the following:
-The CKO and the Pundits
-The Pundits and each other
-The CKO and the two averages
-The Pundits and the two averages
|
|
CKO |
PA |
PB |
PC |
PD |
PE |
PF |
A1 |
A2 |
|
CKO |
1.000 |
0.752 |
0.750 |
0.720 |
0.695 |
0.686 |
0.516 |
0.749 |
0.699 |
|
PA |
0.752 |
1.000 |
0.865 |
0.931 |
0.925 |
0883 |
0.749 |
0.950 |
0.928 |
|
PB |
0.750 |
0.865 |
1.000 |
0.875 |
0.871 |
0.821 |
0.856 |
0.941 |
0.948 |
|
PC |
0.720 |
0.931 |
0.875 |
1.000 |
0.920 |
0.914 |
0.765 |
0.959 |
0.942 |
|
PD |
0.695 |
0.925 |
0.871 |
0.920 |
1.000 |
0.940 |
0.780 |
0.965 |
0.958 |
|
PE |
0.686 |
0.883 |
0.821 |
0.914 |
0.940 |
1.000 |
0.765 |
0.948 |
0.925 |
|
PF |
0.516 |
0.749 |
0.856 |
0.765 |
0.780 |
0.765 |
1.000 |
0.878 |
0.868 |
|
A1 |
0.749 |
0.950 |
0.941 |
0.959 |
0.965 |
0.948 |
0.878 |
1.000 |
0.987 |
|
A2 |
0.699 |
0.928 |
0.948 |
0.942 |
0.958 |
0.925 |
0.868 |
0.987 |
1.000 |
2006
Below
are the rankings produced by the pundits (PA to PI).
A1
is the real average of PA to PI. It reflects how close fighters are to
each other by ignoring the rungs of the ladder. Where a normal ranking
would be 1-2-3, the chart below will show 1- 2.67 - 4.
A2
is the average of PA to PI adjusted to form a ranking where each rank is
given the appearance of being equally spaced. Note that ranks 9 and 10
were tied and thus, for statistical purposes, the rank 9.5 was given to
each.
|
Fighter
|
CKO
|
PA
|
PB
|
PC
|
PD
|
PE
|
PF
|
PG
|
PH
|
PI
|
A1
|
A2
|
|
Akio
Mori |
1
|
5
|
3
|
4
|
3
|
3
|
5
|
4
|
3
|
6
|
4
|
3
|
|
Peter
Aerts |
2
|
2
|
2
|
5
|
2
|
4
|
2
|
2
|
2
|
3
|
2.67
|
2
|
|
Sem
Schilt |
3
|
1
|
1
|
1
|
1
|
1
|
1
|
1
|
1
|
1
|
1
|
1
|
|
Francois
Botha |
4
|
12
|
12
|
13
|
13
|
16
|
14
|
17
|
19
|
17
|
14.78
|
16
|
|
Remy
Bonjaski |
5
|
4
|
7
|
2
|
6
|
2
|
3
|
6
|
6
|
5
|
4.55
|
4
|
|
Siala-Mou
Siliga |
6
|
3
|
6
|
12
|
7
|
9
|
4
|
7
|
5
|
8
|
6.78
|
6
|
|
Ray
Sefo |
7
|
6
|
4
|
6
|
4
|
5
|
7
|
3
|
4
|
9
|
5.33
|
5
|
|
Kaoklai
Kaennorsing |
8
|
7
|
14
|
11
|
14
|
7
|
12
|
8
|
8
|
15
|
10.67
|
11
|
|
Ernesto
Hoost |
9
|
8
|
9
|
3
|
15
|
15
|
8
|
9
|
9
|
13
|
9.89
|
8
|
|
Badr
Hari |
10
|
10
|
10
|
8
|
8
|
8
|
11
|
13
|
14
|
12
|
10.44
|
9.5
|
|
Alexander
Ustinov |
11
|
14
|
13
|
14
|
11
|
6
|
10
|
12
|
12
|
2
|
10.44
|
9.5
|
|
Jerome
LeBanner |
12
|
9
|
5
|
7
|
5
|
11
|
6
|
5
|
10
|
10
|
7.56
|
7
|
|
Nicholas
Pettas |
13
|
15
|
18
|
17
|
16
|
17
|
18
|
18
|
17
|
19
|
17.22
|
19
|
|
Ashwin
Balrak |
14
|
18
|
15
|
15
|
17
|
12
|
19
|
15
|
16
|
11
|
15.33
|
17
|
|
Stefan
Leko |
15
|
11
|
11
|
9
|
9
|
14
|
13
|
14
|
15
|
16
|
12.44
|
12
|
|
Peter
Graham |
16
|
19
|
17
|
10
|
10
|
19
|
17
|
11
|
13
|
4
|
13.33
|
13
|
|
Hong
Man Choi |
17
|
13
|
8
|
16
|
18
|
13
|
16
|
16
|
7
|
20
|
14.11
|
15
|
|
Gokhan
Saki |
18
|
16
|
19
|
18
|
12
|
18
|
20
|
19
|
18
|
14
|
17.11
|
18
|
|
Bjorn
Bregy |
19
|
17
|
16
|
19
|
19
|
10
|
15
|
10
|
11
|
7
|
13.78
|
14
|
|
Mirko
Filipovic |
20
|
20
|