[CQ-Contest] SS Log Checking - What really happens to uniques

[n6tr at teleport.com]

There is some confusion about the log checking process used for the 
SS - in regards to what happens if your cousin gets on and works
you with his indoor dipole and doesn't work anyone else in the 
contest.  

While it is true that this will show up in the SS report as a unique
call - showing a received QSO number of #1, it is NOT true that this
QSO is removed from your score.  The ONLY QSOs removed based upon
the status of the callsign are those that have been judged to be
incorrect.

Perhaps the best way to illustrate this is by showing my report from
the SS CW contest to the group.  There are some interesting things
to note in this report.  First off, I have a higher than average
number of unique QSOs (most with #1) due to efforts made by my host
to have the locals get on and give me a QSO.  In the future, I hope
that we can convince these people to hand out some more QSOs, so
they will appear in the database.

All of these unique QSOs count.

The other thing of note is that I incorrectly put K1TO in San Francisco
and KI7Y (who lives about 10 miles from me) in Orange.  I think a
process improvement here would be to require at least three letters
to be entered for some of the more tricky domestic QTHs: SF, LA, OR.
Perhaps SFO, LOU and ORE.

Please read the explanation of the report at the end.  I had tried to
answer any questions.  If I can make some improvements in the wording, 
please let me know so I can make the changes before the SSB reports are 
available.

One other question I get asked often is "what is a good error rate".

They are all over the map.  First off, there is only one person in
my database who has a golden log - and that is with only two QSOs!
The winning logs in a couple of categories have error rates below
1 pecent - which is excellent.  There aren't many logs below two 
percent.  Anything over 5 percent indicates that you probably could
do much better by trying harder.  It is possible that some of the 
logs over 10 percent are challenged by the CW speeds.

There are many logs in the middle of the pack that are in the 20-40
percent error range.  This is probably the first time these people 
have had their logs checked in this detail.  Hopefully they will 
use this information to become aware of areas of improvement.  

Even at my fairly good error rate, I have come up with several 
actions that I think will help me improve next year.

I have a high amount of pride in the results.  We have had a few people
who had taped the contest go back and review their report against the
log.  In a few cases, we have identified some software improvements
to improve the accuracy.  It is quite possible that there are one or
two things in your report that are wrong - but not more than that.
For next year, we want to expand the information that the checking 
program has available to it to include the time of the QSO and the
band.  This will allow me to automate some of the remaining steps
that I do by hand.  The goal is to automate as much of the process
as possible, so each and every log is checked using the same criteria.

One concept I had to create to do this right were "unstable calls".
These are calls which sent different exchange information at 
different times during the contest (based upon the received information
found in the logs I have).  Without this concept, the log checking
would be full of wrongful deductions.  On CW, there were about 40
calls on this list.

If anyone thinks there is something unfair about how the logs are
scored, it is more likely it is because there is some detail that
wasn't understood (like the above case about uniques being removed).
PLEASE address any concerns you have about it to me or the ARRL.

We have had some very long debates about how this will be received
by the middle of the pack guy.  So far, the feedback that I have 
received from everyone is that this is a good thing.  The ones
who got hit hard are somewhat humbled, but are saying they will 
work on improving next year.

One thing that will be hard for a few people to swallow - it will 
be VERY difficult to do the 79x79 trick.  Of the few who tried,
none of them had a perfect log, and thus, lost some multipliers.

I wonder who will be the first person to turn in a perfect 79x79
log?

So - without furthur ado - here is my W5WMU SS CW report for all
to see.  When you see the final results, you will see that I 
stayed in 3rd place, and that the 4th place score is pretty distant,
so I hope this makes everyone comfortable with "the fox watching
the hen house".  The #2 score ended up 1 QSO ahead of me.

                        SWEEPSTAKES LOG REPORT FOR  W5WMU

Dupe check results
- ------------------
W2ESX is a dupe on line #585.
W6TK is a dupe on line #1039.

Callsign check results
- ----------------------
K5JS was not found in the SS database.  Received QSO# = 1
Calls with same check/section: K5KLA

WB8LXX is a busted call.  The correct call is WD8LXX.

W3WX is a busted call.  The correct call is W2WX.

W2WWP was not found in the SS database.  Received QSO# = 25

N2FW was not found in the SS database.  Received QSO# = 1
Calls with same check/section: WA2YBI

WA9PLT was not found in the SS database.  Received QSO# = 2
Calls with same check/section: N9JF

VE7XO is a busted call.  The correct call is VE7QO.

N5AN was not found in the SS database.  Received QSO# = 1

N5AUO was not found in the SS database.  Received QSO# = 1

KD5EQH was not found in the SS database.  Received QSO# = 2

NG5X was not found in the SS database.  Received QSO# = 1

K5UE was not found in the SS database.  Received QSO# = 1

K5UA was not found in the SS database.  Received QSO# = 1

AD4MZ was not found in the SS database.  Received QSO# = 1
Calls with same check/section: NW6S N4IJ AA4S AD4IE

WU3V was not found in the SS database.  Received QSO# = 1
Calls with same check/section: W5ZDW

K5USL was not found in the SS database.  Received QSO# = 1
Calls with same check/section: W5ZDW

KI5TK is a busted call.  The correct call is KI5TD.

W1EA is a busted call.  The correct call is W1EAT.

N5MEG was not found in the SS database.  Received QSO# = 10

N5KQQ was not found in the SS database.  Received QSO# = 1

KF5EA was not found in the SS database.  Received QSO# = 1

KA5BSE was not found in the SS database.  Received QSO# = 1

KC5BYD was not found in the SS database.  Received QSO# = 1

WD5GJB was not found in the SS database.  Received QSO# = 1

WD5DBV was not found in the SS database.  Received QSO# = 1

KE5FZ was not found in the SS database.  Received QSO# = 1

KC5YOA was not found in the SS database.  Received QSO# = 1

W0MJ was not found in the SS database.  Received QSO# = 1
Calls with same check/section: W5WMU WA5SXU K7RAT

K3HS is a busted call.  The correct call is K3HH.

KZ5D was not found in the SS database.  Received QSO# = 1
Calls with same check/section: K5KLA

There were 6 calls judged to be incorrect.

Precedence, Check and Section Check Results
- -------------------------------------------
QSO #70 KI7Y : A 59 Org should be A 59 Or
QSO #237 W8EO : A 53 Mi should be A 54 Mi
QSO #341 VE3WZ : B 48 On should be A 48 On
QSO #383 K1TO : B 72 Sf should be B 72 SFl
QSO #667 N3KRX : A 90 De should be A 91 De
QSO #703 KF9YL : Q 77 Il should be A 77 Il
QSO #708 N0MA : A 97 Ia should be A 96 Ia
QSO #768 N0VLJ : A 63 Mn should be A 62 Mn
QSO #861 WA9AQN : A 71 Il should be A 61 Il
QSO #878 W6OAT : B 58 Sv should be B 58 Scv
QSO #897 W6VDF : A 57 Org should be A 57 Or
QSO #923 W9SZ : Q 67 Il should be A 67 Il
QSO #1215 KA2QIK : A 72 WNy should be A 82 WNy
QSO #1242 KK7PA : A 75 Org should be A 75 Or
QSO #1260 WA5SXU : B 67 La should be A 67 La
QSO #1389 K8DD : Q 57 Mi should be Q 58 Mi
QSO #1488 K5PQ : A 59 Ar should be A 64 Ar

99.0% of your non dupe QSOs had their exchanges checked.

There were 17 exchange errors found.

Cross check results
- -------------------
QSO #124: Received QSO# 46 should be 36 WA2EYA
QSO #715: Received QSO# 202 should be 204 W0CO

41.8% of your remaining good QSOs were cross checked.

Multiplier Calculation
- ----------------------
List of 79 mults = Eb Sjv Ct Nv NNj Scv Org WWa Bc Sb Em Az Ri NTx Id Ew Wi
Co WNy Oh Mn Or Sf Pac Va WMa SFl Sdg WPa La Mdc Wv Sc On Ep Nc Pr Sd Ak Ab
Lax Nh Me Sv Ok Mi SNj Ky Ut Il ENy Mar Mo Ne Tn NLi Mt In Vi Wy Ga Nd De Al
Ia Ks STx Vt NNy Ms Sk WTx Ar Nm Mb NFl Qc Nl Nwt

Score summary
- -------------
    Raw QSOs = 1514
       Dupes = 2
 Busted QSOs = 25
  Valid QSOs = 1487
Penalty QSOs = 18
  QSO Points = 2938
  Multiplier = 79
- --------------------------
 Final score = 232102
 Error rate =  1.7%

This report was generated by the log checking program.  It was made
possible because you either submitted your log via e-mail or other
electronic format.

Every effort has been made to assure the accuracy of this report.  If
you see something that looks strange, please let us know.  We are
committed to improving the process for future contest.

There aren't any dupe penalties for electronic logs - so if you see
dupes listed, they are simply not counted and there is no penalty.

The callsigns listed after the dupe check are the ones that didn't
make it into the database.  The criteria for a call to show up in
the database is that it must show up in at least 4 of the submitted
electronic logs.  Callsigns that have been judged to be incorrect
will be shown here - along with the correct callsign.  Incorrect
calls are removed from your score, along with a penalty of three
additional QSOs.

Any exchange information (precedence, check or section) that is
incorrect results in that QSO not counting, but no penalties.

Incorrect QSO numbers, or not-in-logs are removed without penalty.

We hope this information will be useful to you as you try to improve
your logging accuracy.

73 and we hope to see you in the next SS!

--
CQ-Contest on WWW:        http://www.contesting.com/_cq-contest/
Administrative requests:  cq-contest-REQUEST at contesting.com