[CQ-Contest] cluster spot analysis

[David Robbins K1TTT]

You can prove anything with statistics...  

Here are the over all top spot getters from all sources for the weekend:

DX		Count
C5Z		491
A61AJ		387
PJ2T		347
VP5B		312
TI5N		309
PZ5A		274
VB2C		250
V26B		250
WP2Z		238
CT9L		237
YV4A		231
HC8N		226
FY5KE		219
D4B		218
HB0/HB9AON	213

Pretty much what might be expected.  The rare operations like C5, A61
come out on top, rare zones like vb2c with full time operation are up
there, Caribbean area and hc8n get lots of spots from stateside if they
are on a lot, etc.

Now consider this comparison showing the count of 'single spotter'
spots.  

These 15 stations got the most spots from people who only made one spot
in the whole contest.  there were 1583 spots that fell into this
catagory out of 38355 i logged in the contest.  These were for 338
different dx stations.  So the average dx station that got spotted by
the 'single spotters' was spotted 4.6 times. This list includes all the
stations that got 10 or more 'single spotter' spots.  The count of
single spots from all sources is what is in the first column below.

The second column shows the number of those spots that were input from
dxsummit.

The third column shows the total spots for each station from all
spotters(the list above).

		single	single	total
DX		World		DXS		world
BY6HY		60		48		83
JT1BV		60		56		94
EA7FTR	33		32		48
A61AJ		27		6		387
RK3MWD	23		23		46
AN7MPM	22		3		89
C5Z		21		7		491
EA1AKS	18		18		33
TM5CRO	17		8		75
SY8A		16		5		115
HB0/HB9AON	11		1		213
CE0Y/SP9PT	11		3		54
RA9JR		11		11		26
VB2C		10		2		250
CT9L		10		3		237

Lots of interesting things here.  Lets take A61AJ and C5Z, the world's
most spotted stations in the contest.  They got somewhere around 5% of
their spots from 'single spotters'.  And of those single spots 33% or
less came from DXSummit.  Now, pick a few others.  BY6HY and JT1BV got
around 2/3's of their spots from stations who didn't spot anyone else in
the whole contest, and 90% or more of those came from DXSummit.  Similar
results apply to ea7ftr, rk3mwd, ea1aks, and ra9jr... which coincidently
if you look at my last message with analysis of ip addresses from
dxsummit, all show up in that list.  

Now, where did an7mpm come from???  they were mostly psk spots, a
relatively low activity mode, non-contest, and special call all combined
to attract some attention, but note, even with a large number of single
spotter spots only a small percentage of them come from dxsummit.
Likewise tm5cro was mostly vhf and iota spots outside the contest.  In
the contest sy8a attracted a fair number of spots as a rare country, as
did the hb0, ce0y, and ct9l, vb2c was in zone 2.  note that all of them
had relatively low percentages of dxsummit single spotters.

One very interesting conclusion from this is that i did not detect any
big new suspects from the cluster data.  Maybe this is a sign that word
is getting around about cheating by spotting yourself on cluster nodes
and all the activity is now on dxsummit.  If this is right then what i
am doing would seem to be working and we just have to keep spreading the
word that dxsummit is no longer a safe haven.  Of course it could be
that the cheaters are getting smarter and found ways to avoid detection,
but somehow I doubt that.


David Robbins K1TTT
e-mail: mailto:k1ttt at arrl.net
web: http://www.k1ttt.net
AR-Cluster node: 145.69MHz or telnet://dxc.k1ttt.net