Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

A fix to the bonus point system

0 views
Skip to first unread message

Peter Ballard

unread,
Feb 6, 2002, 12:57:51 AM2/6/02
to
Seems to me the problem with the bonus point system is that there
is no penalty for *conceding* a bonus point.

Suppose there are always exactly 4 points up for grabs in any one game.
So if a team gets 4 points for the win plus 1 bonus point,
then their opponents lose 1 point.

That would have averted both contentious outcomes in the VB series:
No way would NZ have conceded a bonus point (because it would have
penalised themselves). And SA chased 226 or whatever it was against
Aust to avoid losing a point themselves, rather than merely depriving
Australia of a point.

p.s. I'm not whinging. SAf and NZ were quite within their rights IMHO.

Peter Ballard

Simon

unread,
Feb 6, 2002, 1:19:44 AM2/6/02
to

No, no, you've got it all wrong. The problem with the bonus point is that
it is given automatically if a side achieves an overall RR 1.25 times that
of the opposing team.

What SHOULD happen is that if a side achieves a run-rate 1.25 times that
of the opposing team, then the opposing team must appear on a live
pay-per-view "Wild Nude Bonus Point-off" action match, in which the
opposing team members (all naked) face off against a TV celebrity who
tries to hit them in the bonce with assorted rotten vegetables.

Thus they would only concede a bonus point if they cannot handle the
boncers in the wild nude bonus point-off from of the game. Now that would
be good for ratings.

--
Simon Kent - esrever ni liame

Phil Wise

unread,
Feb 6, 2002, 1:46:41 AM2/6/02
to

"Simon" <ua.ude.tuq@tnek.s> wrote in message
news:Pine.OSF.4.10.102020...@pigeon.qut.edu.au...

>
> No, no, you've got it all wrong. The problem with the bonus point is that
> it is given automatically if a side achieves an overall RR 1.25 times that
> of the opposing team.
>
> What SHOULD happen is that if a side achieves a run-rate 1.25 times that
> of the opposing team, then the opposing team must appear on a live
> pay-per-view "Wild Nude Bonus Point-off" action match, in which the
> opposing team members (all naked) face off against a TV celebrity who
> tries to hit them in the bonce with assorted rotten vegetables.
>
> Thus they would only concede a bonus point if they cannot handle the
> boncers in the wild nude bonus point-off from of the game. Now that would
> be good for ratings.

You mean the nuodo from off the game?

phil

alvey

unread,
Feb 6, 2002, 2:22:07 AM2/6/02
to

"Peter Ballard" <pbal...@ozemail.com.au> wrote in message
news:9d5509fa.0202...@posting.google.com...

> Seems to me the problem with the bonus point system is that there
> is no penalty for *conceding* a bonus point.

Well there is actually. You get beaten by 5 instead of 4.


>
> Suppose there are always exactly 4 points up for grabs in any one game.
> So if a team gets 4 points for the win plus 1 bonus point,
> then their opponents lose 1 point.

This wouldn't avoid the scenario of a side not attempting to win, just
avoiding conceding the BP.

And what happened to the "exactly 4 points...in any one game"? By your
method there could be a six point gap between A and B after one game.


Alvey

Andrew Dunford

unread,
Feb 6, 2002, 2:36:25 AM2/6/02
to

"alvey" <alvey_s...@yahoo.com> wrote in message
news:3c60d...@news.iprimus.com.au...

Just a tiny flaw in the theorem. He meant that a side losing badly is also
*awarded* a bonus point. Thus the gap between the points awarded to each
side is only four and teams are encouraged to capitulate quickly so we can
all get on with something else.

Andrew


Paul Robson

unread,
Feb 6, 2002, 3:02:23 AM2/6/02
to
In article <9d5509fa.0202...@posting.google.com>,
pbal...@ozemail.com.au (Peter Ballard) writes:

>p.s. I'm not whinging

No of course not.

Mike

unread,
Feb 6, 2002, 9:09:25 AM2/6/02
to
"Peter Ballard" <pbal...@ozemail.com.au> wrote in message news:9d5509fa.0202...@posting.google.com...

> Seems to me the problem with the bonus point system is that there


> is no penalty for *conceding* a bonus point.

Of course their is, the winner gains 5 instead of 4 point, this is like saying
a tennis player who loses a rally does not get penalised, he does, his
oponent moves one point ahead of him.

> Suppose there are always exactly 4 points up for grabs in any one game.
> So if a team gets 4 points for the win plus 1 bonus point,
> then their opponents lose 1 point.

This is the same as saying a team who loses should get negative points
eg 2 points to winning side, -2 point to the losing side. Net result would
still be 4 points. You are effectively just making the bonus 2 points instead
of one point.

> That would have averted both contentious outcomes in the VB series:

The only contentious thing is that the Aus are not happy they were not
up to the competition.

No matter what the rules, some form of method must be used to separate
teams that end up level pegging, and this will always allow for some
idiosyncrasies.

Cheers
Mike


Michael Jennings

unread,
Feb 6, 2002, 9:28:22 AM2/6/02
to
"alvey" <alvey_s...@yahoo.com> wrote in message news:<3c60d...@news.iprimus.com.au>...
> "Peter Ballard" <pbal...@ozemail.com.au> wrote in message
> news:9d5509fa.0202...@posting.google.com...
> > Seems to me the problem with the bonus point system is that there
> > is no penalty for *conceding* a bonus point.
>
> Well there is actually. You get beaten by 5 instead of 4.
> >
> > Suppose there are always exactly 4 points up for grabs in any one game.
> > So if a team gets 4 points for the win plus 1 bonus point,
> > then their opponents lose 1 point.
>
> This wouldn't avoid the scenario of a side not attempting to win, just
> avoiding conceding the BP.

However, once they have avoided conceding the bonus point, they have
nothing more to lose, so they might as well try to win. A situation in
which a side is better off in the competition by playing worse, which
happened in the SA-NZ game, is eliminated. I think this is a big
impovement.

>
> And what happened to the "exactly 4 points...in any one game"? By your
> method there could be a six point gap between A and B after one game.
>

Yes, but if one side scores five points, and the other loses
one, then the total number of points awarded in the competition
increases by four. This constant number of total points awarded in the
competition being constant does avoid a number of potential problems.

Actually, there is a possibly easier to understand way of doing the
same thing. Award zero points for a loss in which the other side
scores a bonus point, one point for a loss in which the other side
does not score a bonus point, five points for an ordinary win, and six
points for a win including a bonus point. This is mathematically
equivalent to what is described above (in that it results in sides
ending up with the same places on the table by the same margins) but
avoids the confusion of negative points. (Or you could award zero,
one, four, and five, which isn't quite mathematically equivalent but
is close, and still avoids the same problems).

Michael.

Gary Williams

unread,
Feb 6, 2002, 2:57:12 PM2/6/02
to
"Mike" <nos...@btclick.com> wrote in message news:<xMa88.31776$oK1.298980@NewsReader>...

> "Peter Ballard" <pbal...@ozemail.com.au> wrote in message news:9d5509fa.0202...@posting.google.com...
>
> > Seems to me the problem with the bonus point system is that there
> > is no penalty for *conceding* a bonus point.
>
> Of course their is, the winner gains 5 instead of 4 point, this is like saying
> a tennis player who loses a rally does not get penalised, he does, his
> oponent moves one point ahead of him.

But your logic only applies to a scenario with exactly two opponents.
The problem is that the opponent _also_ moves ahead relative to
contenders who are not involved in the match. In the present
instance, NZ giving up the bonus point to SA moved SA ahead not only
of NZ, but also of AU. And since what NZ needed to do was not finish
first but avoid finishing third, surrendering the bonus point to SA
cost it nothing. However, imposition of a penalty point would have
cost NZ relative to AU, and made the tactic unrewarding.

Well, in a competition like this you gotta have a tiebreaker. But net
run rate, being less "lumpy", is harder to manipulate than bonus
points. On the other hand, it's a little harder to compute rapidly in
one's head, so that even if the fans can do it, the announcers may not
be able to handle it.

The underlying problem comes from choosing finalists from a round
robin competition.* If the competition for second place is close, the
first place team may have an opportunity to choose its final round
opponent. Say that going into the final game, between A and B, the
standings are A:20, B:12, C:12. If A prefers to play B in the final,
they merely let B win.

But the bonus point system creates more situations in which this kind
of manipulation can be brought off. If the standings are A:24, B:8,
C:12, then A's loss to B will not put B in the final unless B also
finishes ahead of C on the tiebreaker, something A may not be able to
manipulate; at least not by enough at this stage to make a difference.
But if there are bonus points, then A concedes both the match and the
bonus point, moving B ahead of C, all completely within A's control.


*It's not just cricket, and it's not just bonus points. We've had a
scandal recently in which a speed skater, having secured his position
on our Olympic team, skated his final race badly enough to finish
behind a close friend, giving the friend a point in the standings that
was enough to move him past another skater for the final place in the
team.

Gary Williams

Michael Jennings

unread,
Feb 6, 2002, 4:09:23 PM2/6/02
to
"Mike" <nos...@btclick.com> wrote in message news:<xMa88.31776$oK1.298980@NewsReader>...
> "Peter Ballard" <pbal...@ozemail.com.au> wrote in message news:9d5509fa.0202...@posting.google.com...
>
> > Seems to me the problem with the bonus point system is that there
> > is no penalty for *conceding* a bonus point.
>
> Of course their is, the winner gains 5 instead of 4 point, this is like saying
> a tennis player who loses a rally does not get penalised, he does, his
> oponent moves one point ahead of him.

Yes, but that is a two player game. Ultimately one of the two players
much win. In a competition table with bonus points and more than one
team, a third team may win, and the losing team in the present
situation does not get penalised relative to the third team. It was
this lack of a penalty against a third team that New Zealand took
advantage of.


>
> > Suppose there are always exactly 4 points up for grabs in any one game.
> > So if a team gets 4 points for the win plus 1 bonus point,
> > then their opponents lose 1 point.
>
> This is the same as saying a team who loses should get negative points
> eg 2 points to winning side, -2 point to the losing side. Net result would
> still be 4 points. You are effectively just making the bonus 2 points instead
> of one point.

No he isn't. Making the bonus point 2 points instead of 1 would still
not penalise the losing side relative to a third team. It would also
not have affected the outcome of this competition. However, adopting
Peter's proposal, or the very similar one I have made elsewhere in
this thread would have made things difference.


>
> > That would have averted both contentious outcomes in the VB series:
>
> The only contentious thing is that the Aus are not happy they were not
> up to the competition.

Australia didn't make the finals because they didn't play very well: I
don't think many people are disputing this. (However, neither of the
other teams played much better). What we are questioning is a system
in which New Zealand qualified for the finals but would not have done
so if they scored more runs in their last game.


>
> No matter what the rules, some form of method must be used to separate
> teams that end up level pegging, and this will always allow for some
> idiosyncrasies.
>

Yes, and I think it is better to use a simple system than a
complicated one. I think net run rate on its owns serves us fine, and
leads to fewer idosyncracies than this overcomplicated system of bonus
points then head to head results then net run rate. (Net run rate can
lead to teams batting slowly in order to affect who their opponent may
be, as Australia tried unsuccessfully to do in the Super Six stage of
the last World Cup. However, it is never going to lead to a situation
where batting slowly will improve their own chances of going through.

Michael.

Gary Williams

unread,
Feb 6, 2002, 9:09:54 PM2/6/02
to
mj...@amtp.cam.ac.uk (Michael Jennings) wrote in message news:<581e1f6f.0202...@posting.google.com>...

> "Mike" <nos...@btclick.com> wrote in message news:<xMa88.31776$oK1.298980@NewsReader>...

> > No matter what the rules, some form of method must be used to separate


> > teams that end up level pegging, and this will always allow for some
> > idiosyncrasies.
> >
> Yes, and I think it is better to use a simple system than a
> complicated one.

There is the _real_ fallacy...that there is a final in which a team
that has already been outplayed by an opponent gets a second shot
after failing in the round robin. Any system of
round-robin-plus-final may lead to manipulated results in the
preliminary rounds, and to a superior team being upset by a weaker one
in a short series final. Of course, if there _weren't_ a final to
play toward, the paying public might lose interest as soon as one team
outdistanced the field. But which is more important--determining the
best team, or keeping the turnstiles clicking?

Oh...obvious answer....

Gary Williams

0 new messages