From time to time WSPR's reports a decode that is just "Too good to be true".
To a newcomer they are something to get excited about. Is this WSPR mode really that amazing.
Sometimes they just "don't look or feel right" to an experienced WSPR's eye.
Sometimes they are just plain "WRONG"

If it fails the Database filters, IT thinks it is Suspect and is simply not entered in the DB.
If it passes the Database filters and is entered it may still be suspect.
Filters are not intuitive, they don't ask all the right questions and think about the Answers.

There are a few simple checks you can do, that should confirm if a Decoded Spot is Suspect.

Date and Time are rarely suspect, unless you know you were not on air at that time.

Look at the Callsign, Is it a Normal Format (prefix, number, suffix), recognise the prefix?
No, a quick check with QRZ.com or a similar DX Call book Listing.
"The search for 6p7frs produced no results!, MMM, 6p7frs looks like a callsign from Pakistan".
Not a good result, Suspect ?

Look at the Band/Distance, 50MHz, 5123kM, a Long way to go on a quite band, 1600kM is my best DX today. Not a good result Suspect?

Look at the S/N ratio -29dB, thats right at the decoders limits. We know WSPR was working very hard trying to decode something, ALL_WSPR.txt confirms the 6p7frs data looped through the error correction system 16,018 times to produce this Decode. Not a real good sign!

Look at the Grid Square BG24 the WSPR Map tells me in the middle of Antartica, Only 4 digits, A Pakistani in Antartica far from all known bases. Not an expected result Suspect?

Look at the Power, 501 Watts, Odd figures, Wattmeter that accurate?, Not really QRP,
(10 Watts or Less). Again Not a good looking result Suspect?

So our Suspect Spot Below Failed on most of the Simple Tests and I Believe it should be consigned to the Bit Bucket where it obviously belongs.

Timestamp Call MHz SNR Grid Pwr Reporter RGrid km az
2012-01-21 00:44 6P7FJS 50.294520 -29 BG24 501 VK4ZBV QG62ml 5132 255

Most of these type of suspects result from WSPR trying to decode interference, spectral lines which can be seen on the WSPR waterfall as a horizontal line often extending full screen width, little or no obvious frequency drift, AC Mains harmonics, Switch mode power supplies.

Another cause of Suspect Spots becomes obvious from the above:
If a stream of received data is cycling around the error correction system up to 19,999 times to obtain a Decode before discarding it, which depending on your computer speed may take up to minutes to resolve, and you change band or frequency hop to another band during that time the reported decode will obviously be reported on the wrong band.

I hope this information will assist those both Old and New to WSPR's to spot suspect spots which should be reported to the WSPRnet Forum thread "WSPRnet.org Site" for removal from the database.

73 Paul VK4ZBV.