Hi Probonopublico,
The general consensus on this seems to be that although the
Duckworth-Lewis model is more statistically sound than most
alternative methods (e.g., "average run rate"), it is not
statistically sound under extreme circumstances. For example, when a
team's performance has been far above average, Duckworth-Lewis tends
to produce unrealistic results.
Even Messrs. Duckworth and Lewis acknowledge that the basic
assumptions of their method only apply to about 95% of cases. In other
words, the model produces unrealistic results in as many as 5% of
cases.
"The use of the simplifying single table of resource percentages means
that actual performance must necessarily be assumed to be proportional
to average performance. In 95% of cases this assumption is valid, but
the assumption breaks down when an actual performance is far above the
average,"
Source: Cricket.org: DL FAQ
http://www.cricket.org/link_to_database/ABOUT_CRICKET/RAIN_RULES/DL_FAQ.html
Additional sources:
"The problem with Duckworth-Lewis...
...the method fails to acknowledge the 'regression toward the means'
effect...
The relative resources table only makes sense for average
performances. After a bad start we expect a team to recover and after
a great start we expect that there could be a few hiccups along the
way."
Source: Melbourne University Mathematics & Statistics Society: Paradox
http://www.ms.unimelb.edu.au/~paradox/issues/p99-2.pdf
"There have been over 200 applications of the D/L method in the last
four years, and I can recall only a handful of cases where the target
has been seriously questioned. This is excellent by any yardstick, but
we could still argue! For example, if the 1999 India vs Sri Lanka
World Cup match had terminated with Sri Lanka at 117 for no loss in 25
overs, in reply to India's massive 371, Sri Lanka would have won
according to the D/L method!"
Source: "Is Jayadevan's proposed method better than the
Duckworth/Lewis method?" by Srinivas Bhogle, hosted by cmmacs.ernet.in
http://www.cmmacs.ernet.in/nal/pages/dlj.html
"[T]he D/L method is not perfect. Srinivas Bhogle of the National
Aerospace Laboratories (NAL), Bangalore, brother of the commentator
Harsh Bhogle, has studied the D/L method extensively. He said that it
has its 'blind spots'.
....
Another weakness of the D/L method is that it allows the team batting
second, Team 2, to play according to a strategy in situations when the
target is large but there is the prospect of rain curtailing the
match. All that the team has to do is keep its wickets intact, even if
the scoring rate is less than required."
Source: "For a Fair Formula," Frontline magazine
http://www.frontlineonnet.com/fl1924/stories/20021206004410400.htm
"..They decide that their best hope lies in exploiting a well-known
anomaly of the D/L method which strongly favours the team batting
second, especially till the 30th over or so, if the team holds on to
its wickets (loses less than 2-3 wickets).
Of course, D/L comes into play only after 25 overs are completed. So
the wise thing to do is to ensure that England is best placed at the
25-over mark. The next best thing to do is to ensure that no more than
30 overs are bowled (after 30 overs, D/L becomes 'fair' again)."
Source: Rediff Cricket: "Rain Saves India"
http://www.rediff.com/cricket/2002/jul/06rain.htm
"Everyone knows the D & L method produces some anomalies, sometimes
producing results that go against the likely natural outcome," hosted
by espnstar.com:
http://www.espnstar.com/jsp/cda/studio/id=87820&aid=489167&ecode='COL3_LP'&colid=398678studio_pastcoldetail.html
Further reading:
"The dummy's guide to Duckworth-Lewis" Dr Srinivas Bhogle, hosted by
cmmacs.ernet.in:
http://www.cmmacs.ernet.in/~swathi/dl.html
"Hooper calls for change in Duckworth-Lewis method," hosted by
hindu.com:
http://www.hindu.com/thehindu/2002/06/14/stories/2002061405592000.htm
"Duckworth-Lewis system to change," hosted by cricket-online.org:
http://www.cricket-online.org/news/archive/2001/August/12_AUG_2001_TWALSH2.html
"How we sold Duckworth-Lewis" by Tony Lewis, hosted by orsoc.org.uk:
http://www.orsoc.org.uk/about/topic/news/duck.htm
search strategy: "duckworth lewis", flaw, flaws, problem, problems,
anomaly, anomalies, statistics, statistically
I hope this helps. |
