Self-assessment: LCR (Log Check Report)

The LCR (log check report) for the 2025 CQ WW SSB contest have been distributed. This is an opportunity to see what specific mistakes were made. I know, that can be painful! Nevertheless it is a mandatory exercise if you want to improve. I set a personal error rate target of less than 1%. That isn't easy when operating SO2R and 2BSIQ.

It may be of interest to learn how I read these reports and use them. I'll take you through my analysis. Other contesters may draw different conclusions or focus on different items. Either way, there is more to the analysis than thinking "oops!" when you look at each error. Understanding how errors are made can lead to future improvement. We go to a lot of trouble to make a big score and the competition is fierce, so every point removed from the log is worth attention.

The LCR for this contest is a little different since it was a multi-op (M/2) with 5 operators, including myself. Due to the high rates and perhaps the inclusion of less experienced operators I was expecting a higher than usual error rate. It was a pleasant surprise to learn that we did pretty well despite my worries.

The Cabrillo does not contain the operator call signs. This can be found in the logging software database and in the ADIF that is usually exported after the contest for upload to LoTW and other applications. I deliberately did not make the association, leaving it to each operator to do it themselves. It is an individual responsibility to self assess. There are multi-op teams that produce reports for each operator.

************************** Summary ***************************
    7999 Raw    QSO before checking (does not include duplicates or missing exchanges)
    7889 Final  QSO after  checking reductions

   21766 Raw    QSO points
   21005 Final  QSO points

     170 Raw    zones
     170 Final  zones

     626 Raw    countries
     625 Final  countries

     796 Raw    mults
     795 Final  mults

17325736 Raw    score
16698975 Final  score

    1.4% error rate based on raw and final qso counts
    3.6% score reduction
     30  (0.4%) not in log
     60  (0.8%) incorrect calls
     21  (0.3%) incorrect exchanges
      0  (0.0%) missing exchanges
    175  (2.2%) duplicates removed
     24  (0.3%) calls unique to this log only (not removed)

Logging and log checking is not an exact science. Our log showed 8003 QSOs yet 4 seem to have disappeared. It's a curiosity that is negligible so I didn't delve into it. The 1.4% error rate is acceptable considering the team and circumstances. We only lost one multiplier. So far so good.

Not in log (NIL)

You work a station yet you don't show up in their log. What is going on? There are several possibilities, some your fault and some the fault of the other operator.

• Optimism: Contesters are optimistic. That can lead to over-enthusiasm as we hover over the keyboard to log that new multiplier or hard won contact. QRM, QRN, did they really copy me? Despite any lingering doubts, we log the contact. It's worth taking a few seconds to get positive confirmation that the contact was logged by the other station.
• Overlap: On a crowded band it is surprisingly easy to believe that you worked a station that you did not. This happens more often while running since the other station doesn't send your call. If another station in your skip zone starts running on your frequency you may not notice at first. Doubt may arise when the other station's timing is off from your own. Delays and overlap can also be due to SO2R, while waiting for a transmission to end on the other QSO. If it happens twice in a row, listen closely. Someone else may be there. 
• OOB: This is a problem that is mostly only experienced by those of us working Americans. Here is an example from the LCR:
  
   7099 PH 2025-10-26 0015 VE3VN     4   K*****   04 
   7099 PH 2025-10-26 0032 VE3VN     4   K****    05 
  
  Do you see what happened? How about if I explain that OOB stands for out of band? Outside the US most countries, including mine, have no regulated mode sub-bands. We abide by voluntary band plans. It is easy to click on a spot and not notice the frequency. We don't know the details of US regulations and we don't know where various license classes are permitted. We might take the time to ask if they are indeed OOB, or log them and keep running to save time, just like for dupes. Later they scrub their logs to erase the evidence of their transgression. I obscured the call signs in the LCR extract to protect the guilty.
• Oops: It is quite easy in the heat of a contest to forget to log a QSO when hunting at high speed. You type in the data and spin the VFO or click a spot to move on to the next. But you didn't press enter and the QSO wasn't logged. You soon forget the call sign and see it again, so you try to work them. They may tell you "b4" or "dupe" but you say no and try to complete the contact. Now it's the other station making a mistake and they will get a NIL if they won't work you. That's another good reason to work dupes and avoid time wasting arguments.

Modern log checking software is able to detect copying errors and still correctly match QSOs. That is rarely the reason for a NIL. You're not in the other station's log and you must try to do better.

I have to admit that NIL errors annoy me since they are largely avoidable. I've discussed them in this blog going back many years, here, here and here to give a few examples. I am not exempt from my own advice to do better.

Incorrect calls

It is easy to get a call wrong, either by a copying error or typo. If the other station is running and doesn't send their call every QSO or two you might not notice the error. Human and skimmer spots are also frequently wrong, which is why it is highly advisable to rely on your ears before logging the contact.

 14121 PH 2025-10-26 1954 VE3VN     4   HB9      14  incorrect call 

While running it is quite common for the other station to not really listen to you announcing their call or even caring whether you said it correctly. You get the penalty, not them. By not offering a correction they save precious time. It is your responsibility to confirm that the call sign is valid. In case of doubt make it clear to the other station that they are not being logged until they confirm their call. Do it explicitly or by sending their call with a "?" (on CW) to elicit a repeat.

An important tool to assess the validity of a call sign is the Check Partial window of the logging software (N1MM, DXLog, etc.). It will show matches and near misses to the SCP database (super check partial), call history and your current contest log. If you see a question mark instead of a check mark next to the call ask the station for confirmation. Non-contesters who call you often won't get a check mark, so verify. Even you are certain that you copied the call correctly they may have made a mistake. This can easily occur when sending CW by hand.

Unique calls are variants of incorrect calls. These are calls that are not found in other participants' logs. That's suspicious since it's very unlikely that you'd work a station that no one else has. But it does happen if you have a big signal and there isn't much action. Non-contesters will call you to say hello and try to help with your score. I see a lot of these callers on 40 meters phone when CQ'ing above 7200 kHz. If you sound friendly and welcoming casual ops are more likely to call. 

Even so it is likely that many uniques are indeed calls incorrectly copied. There's no easy way to tell so if your log hasn't been artificially stuffed with them you'll be allowed credit. There are contests in which these contacts are removed from your log without penalty. Again, the lesson is to be certain, so verify.

Incorrect exchanges

For the majority of contacts CQ WW there is really no exchange to log. The zone is determined by the call sign, and logging software will pre-fill it for you. Call history will pre-fill the exchange in cases where the zone cannot be derived from the call sign. However, it is your responsibility to verify the pre-filled exchange.

US calls are the greatest risk. Hams in that country can keep their calls when they move to a different district. Few contests decree that they sign portable; e.g. KX9XXX/6. Whether or not the correct zone is pre-filled, you must check the sent exchange. Enter the correct exchange if it differs from what was pre-filled. 

Many contesters become negligent when they get tired in a long contest or they lose focus while operating 2BSIQ. But there are no acceptable excuses for exchange errors.

 7086 PH 2025-10-25 0005 VE3VN     4   VE2DXY   05  correct     2
 7086 PH 2025-10-25 0014 VE3VN     4   S51CK    14  correct    15
 7086 PH 2025-10-25 0027 VE3VN     4   YP3A     15  correct    20
 7220 PH 2025-10-25 0120 VE3VN     4   K4RXH    08  correct     5 

A common error is to forget to correct the exchange after making and correcting a call sign typo. Depending on the logging application the exchange may not be automatically updated after correcting the call. 

Unless the exchange is very long (e.g. ARRL Sweepstakes) it is usually faster and more accurate to wipe the QSO and reenter the entire call. The corrected call sign then pulls up the correct exchange to pre-fill. Tabbing back and forth when running a pile up, especially while SO2R, is a guarantee of busted exchanges. 

Duplicates and the other guy's mistakes

Back in days of yore when logging wasn't done by computer and we actually sent call signs and exchanges by hand, duplicate contacts were penalized. They had to be found and removed from the log either during the contest (dupe sheets, anyone?) or after. It was a tedious process that was eventually automated by computer logging.

Contest sponsors eventually saw the light and recognized the value of leaving dupes in the log. It made log cross-checking easier and more accurate and put an end to the practice of on-air arguments about whether we've worked before. Dupes were henceforth removed without penalty. Indeed you were encouraged to leave dupes in the log. However, some never lost the habit of scrubbing their logs.

Always always always work dupes. Don't waste time arguing. Perhaps the other station miscopied your call on the current or previous QSO. By not working them again you risk a NIL. 

We had a 2% dupe rate. That's high but not unusually high. It's bound to happen with 8000 contacts in a single contest. Our LCR showed how often our call got mangled. Most were either typos or poor copying, while a minority were due to bad spots. 

 ************* Stations Copying VE3VN Incorrectly *************

 14156 PH 2025-10-25 0000 ***ON    11   VE3VL    04 
 14156 PH 2025-10-25 0001 ***LY    20   VE3WN    04 
 14000 PH 2025-10-25 0054 ***PLL   11   VE3VL    04 
  7000 PH 2025-10-25 0342 ***FKW   15   VE3VW    04 
  7070 PH 2025-10-25 0406 ***DOF   15   VE3WN    04  

Do not simply ignore this section of the LCR. I study dupes to think of ways to improve how we enunciate our call sign, choice of phonetics, CW speed, etc. Every mistake, ours or theirs, is an opportunity to learn and improve. The same goes for busted calls and exchanges. How can you make it easier for others to log your call and exchange more accurately. Again, calls obfuscated in the LCR extract above.

Changing your operating practices can result in fewer dupes. That frees up time to work new stations and earn more points.

How to improve?

Perusing the LCR there are several lessons to me and my team that can lead to improvement in future contests:

• Watch for bad spots! They'll appear on the band map exactly at your operating frequency. Don't simply shrug and plow onward. Send your call more often and more slowly so that stations drawn to the bad spot will recognize the mistake and not waste their time or yours by logging dupes.
• Operators may accidentally speak their own call sign instead of the one used for the multi-op station. Train yourself and your ops to used the pre-recorded messages. For some reason in phone contests many operators unnecessarily speak rather than push a function key. I like to save my voice and make fewer mistakes. Accurate, clear messages are superior to scratchy and negligent voicing when tired.
• QSY when a bad spot draws too many dupes. We are loathe to abandon productive run frequencies yet sometimes we must to avoid wasting time on dupes, QRM and other annoyances. With spectrum displays it doesn't take long to find a new run frequency even on the busiest bands during a popular contest.
• Never react to the pile up. There are more stations waiting in the wings and you don't want them to leave if you take too long to get to them, but that is never an excuse for busted calls and exchanges. Take the time to get it right, and ask for repeats if necessary. With practice you'll get better and the rate will naturally rise. Don't force it. Accuracy comes before speed.

There are contests where QSOs are deleted from both logs if either operator made an error. That can be helpful since it makes operators more careful. I sent your call incorrectly? You have an incentive to try again rather than shrug and let me lose the points. I have mixed feelings about those contests. Rather than dive into that here I'll direct you to an article I wrote about it several years ago.

One feature you should become familiar with is X-QSO. If you are uncomfortable with the accuracy of a call sign or exchange and you are unable to quiz the other station, or you're not sure they logged you, you have the option of marking the contact using the X-QSO feature. 

By marking the contact as an X-QSO you are claiming no credit for the QSO and will not be assessed a penalty (if there is an error) while allowing the other station to be awarded points if the QSO is correct in their log. The most popular contest logging applications, N1MM and DXLog, support this feature. Marking the contact is a judgment call. Make your decision quickly and move on.

One final note about log analysis in a multi-op. Most contesters want to learn from their mistakes, but not everyone. I gave the operators all the data they need to correlate errors to their activity. I leave the rest to them. My primary objective was for everybody to have fun making a big score with a big station. In that we succeeded, spectacularly.