Winch Ester System Press Release

Waratahs cheated out of semi-final spot by inaccurate scoring system Niven Winchester ([email protected]) Bonus points distort league tables. In the recently completed Super 14 round robin, the eightwin, one-draw Crusaders qualified for the semi-finals at the expense of the nine-win Waratahs. Most points systems in sports other than rugby union would rank the Waratahs ahead of the Crusaders. However, the Crusaders earned seven bonuses to the Waratahs’ five, allowing the Christchurch-based team to draw level on competition points and ultimately squeeze the Waratahs out of the semi-finals by virtue of a superior points differential. Does the allocation of Super 14 competition points ensure an accurate ranking of teams from best to worst? This question was addressed in an article recently published in the Journal of Quantitative Analysis in Sport, a peer-reviewed academic outlet published by the Berkley Electronic Press.1 Using mathematical optimisation, the research found that it is too easy for weak teams to earn a four-try bonus. Indeed, to the extent that a dominant team takes its foot of the pedal after establishing a significant lead, being weak can make it easier for a team to earn a four-try bonus. The Red’s round seven 26-50 loss to the Chiefs, which netted the Queenslanders a four-try bonus, is a case in point. The study also found that narrow-loss bonuses are handed out too eagerly. After all, a loss by seven or fewer points enables a losing team to gain such a bonus even though they required a converted try just to tie the game. If the sole intention of bonuses is to rank teams according to strength, the research concluded that bonuses should be awarded for scoring two or more net tries (tries scored minus tries conceded) and losing by five or fewer points. In addition to withstanding mathematical scrutiny, a net-try bonus is preferred to the current try bonus as it is a function of teams’ attacking and defensive abilities, rather than just attacking prowess. The criterion for a narrow-loss bonus also seems more sensible than that currently used – a losing team is required to be within a converted try or two penalties of winning to earn this bonus. Table 1 displays 2009 Super 14 standings when optimal bonus points are employed. The Waratahs would have earned six bonuses and the Crusaders two under the optimal system, resulting in a fourth-place finish for the Waratahs and a sixth-place ranking for the Crusaders. Ultimately, the Waratahs should be playing in Pretoria next week while the Crusaders prepare for next season. The use of non-optimal bonus points in the Super 14 and the Tri-Nations may partially explain Australia’s and New Zealand’s quarter-final defeats to England and France, respectively, at the 2007 Rugby World Cup. Specifically, several rugby experts, including then Australian coach John Connolly, suggested that attacking styles adopted by the two Southern Hemisphere teams was partly responsible for the unexpected results. In turn, habits garnered by Australian and New Zealand players are likely to be influenced by the frivolous four-try bonus point, which is not offered in the Six Nations.                                                              1  Winchester, N. (2008), “Shifting the ‘goal posts’: Optimizing the allocation of competition points for sporting contests” Journal of Quantitative Analysis http://www.bepress.com/jqas/vol4/iss4/1.   

in

Sports,

2008,

4,

Article

1.

Available

at:

TABLE 1: OPTIMAL SUPER 14 STANDINGS, 2009

1

1

Bulls

10

0

3

2

PNTS DIFF. 67

2

2

Chiefs

9

0

5

1

102

42

3

3

Hurricanes

9

0

4

2

101

42

5

4

Waratahs

9

0

4

2

29

42

6

5

Sharks

8

0

3

3

43

38

4

6

Crusaders

8

1

1

1

33

36

7

7

Brumbies

8

0

2

1

6

35

8

8

Force

6

1

3

3

53

32

9

9

Blues

5

0

3

3

-30

26

10

10

Stormers

5

0

1

3

-14

24

11

11

Highlanders

4

0

2

5

-15

23

13

12

Reds

3

0

2

3

-122

17

12

13

Lions

4

0

0

1

-125

17

14

14

Cheetahs

2

0

1

0

-128

9

PLACE1 PLACE2 TEAM

WIN

DRAW

B1

B2

COMP PNTS 45

Note: PLACE1 and PLACE2 denote seedings under the current allocation of competition points and the optimal allocation of competition points; B1 is awarded for scoring two or more net tries; and B2 is given for losing by five points or fewer.