Re: quite an achievement

"Robert Henderson" <philip@xxxxxxxxxxxxxxxxxxxx> wrote in message
news:b5HW7OEfWjgHFwWw@xxxxxxxxxxxxxxxxxxxxxxx
In message <5uepc3F1hgmuoU1@xxxxxxxxxxxxxxxxxx>, David North
<dnorth@xxxxxxxxxxxxxxxxxxxxxxxxxxx> writes

You miss the point, as usual. If you're looking for the
*average* standard
ROTFL! The discussion was about the overall quality of the players
then
and now. RH
What innumerate and irrelevant distinction do you intend to
draw between the "overall quality" and the "average standard"?
Oh dear, what a ricket for a professional mathematician. The average is
just that, an average which can hide a multitude of sins in terms of
difference within the population being averaged.

And it didn't take long for the "translations" to appear.
"Average *standard*", not "average. The proposal wasn't that the
population be averaged, but that we look at the standard of the
typical member of that population, rather than at the standard of
unusual members
. For the present purposes, the statistical average
is not relevant; we already know the RPW, and it tells us nothing
about the quality of the players. What we need is some *other*
evidence about players who achieved roughly that RPW, whether as
batsmen or bowlers, that we can compare across eras.

The overall standard
can be expressed in various ways, for example, the number on or near
the
median,

I would take bets that the number exceeding the median was
close to 50% both in 1955 and 2007, and that the number on or near
the median was very small both years.

Right. Off you go to test your claim. RH

the number who exceed the median by twenty per cent and so. Rh

That's a measure of dispersion. What are you claiming this
statistic would tell us about the overall standard, and why?

Easy. The overall standard can be assessed by referring to things such
as
how many players are well above average or median, how many around the
median, how many well below the median. RH

OK, here are the percentages, for three seasons, of bowlers with averages
within 20% of the median bowling average for that season, < 80% of the
median and > 120% of the median. In each case, I have excluded bowlers who
bowled less than 50 overs.

Season A
<80% - 25.6%
Within 20% - 33.1%
120% - 41.3%

Season B
<80% - 29.8%
Within 20% - 25.8%
120% - 44.4%

Season C
<80% - 30.8%
Within 20% - 26.9%
120% - 42.3%

Tell us what those figures tell you about the overall standard in each
season.
--

A single season tells you nothing. You need a series to evaluate a trend.
RH

So if I give you the same figures for three sequences of five seasons, do
you think that will tell you anything?
--
David North
Test Career Ratings: http://www.lanefarm.plus.com/cricket/ratings


.

Relevant Pages

  • Re: quite an achievement
    ... Where is the evidence that playing with outstanding players makes ... but that we look at the standard of the ... I would take bets that the number exceeding the median was ... OK, here are the percentages, for three seasons, of bowlers with averages ...
    (uk.sport.cricket)
  • Re: quite an achievement
    ... But not a specialist batsman. ... The overall standard can be assessed by referring to things such ... the median, how many well below the median. ... is a sign of greater quality than 5 players averaging 40+ and 80 players ...
    (uk.sport.cricket)
  • Re: Using standard deviation to make predictions
    ... > I understand and I can calculate mean, median, variance and standard ... I've read that one can make predictions about data using ... > standard deviation. ...
    (sci.stat.math)
  • Re: quite an achievement
    ... The discussion was about the overall quality of the players ... but that we look at the standard of the ... the median was very small both years. ... OK, here are the percentages, for three seasons, of bowlers with averages within 20% of the median bowling average for that season, 120% of the median. ...
    (uk.sport.cricket)
  • Re: random numbers from a lognormal distribution
    ... >Shouldn't that be "If X is a standard normal variate then ... and standard deviation from the normal distribution to the lognormal ... functions, and for any non-decreasing function g, the percentiles ... In particular, if m is the median of a normal distribution, then ln ...
    (sci.stat.math)