I've always thought that a single large public file representing the
merged data from all submitted logs would offer all kinds of interesting
statistics ... some fairly simple scripts could probe the data for
almost anything. All of the logs are there, and the folks who manage
pileup.ru have obviously compiled their own master database anyway.
Actual callsigns wouldn't even have to be used ... each callsign could
be converted to a unique entry number and treated as text.
73,
Dave AB7E
On 4/1/2013 6:02 PM, Ward Silver wrote:
By publishing the number of hours worked with the score, using
whatever time-on calculation the log-checkers feel is appropriate for
that contest, it would be straightforward for an interested
third-party to calculate score/hr statistics.
Public, validated logs could also be parsed for all sorts of
time-related stats:
- best first 24 hour score
- best second 24 hour score
- best N hour score
- fastest to N points
and so on. It would be the equivalent of baseball's sabermetrics -
what is the analog of "slugging percentage"?
If the data was there and someone cared about the calculation, it
would probably be performed. We might find an interesting way to
compete that doesn't require a new formal category. Or we might
decide that what we already have is good enough and that more
categories don't really change things. Nevertheless, the impact on
the contest sponsors, who have very limited resources, would be minimal.
73, Ward N0AX
_______________________________________________
CQ-Contest mailing list
CQ-Contest@contesting.com
http://lists.contesting.com/mailman/listinfo/cq-contest
|