Wurdella field guide
Keep a Word-Game Mistake Log You Will Actually Read
Nobody needs a spreadsheet full of every word they have ever guessed. A useful mistake log is smaller and slightly less flattering: it records the decision you would change, the evidence you had at the time and one thing to try next time. Ten honest entries can be easier to act on than a month of unexplained scores.
Give the log one job
The log's job is to reveal a recurring decision, not to produce a detailed diary of a session. Use one line per round you want to review. Record the date, difficulty, outcome, the row you would change and a short reason. Leave out the full board unless a specific tile arrangement is necessary to understand the mistake later.
A compact entry might read: 'Default; lost; row four; returned yellow A to the same square; mark rejected positions before typing.' That line contains a situation, an action and a replacement habit. 'I am terrible at vowels' contains none of those, however convincing it feels after a bad round.
Separate three kinds of trouble
Use three labels: evidence, choice and vocabulary. Evidence means you overlooked something the board already told you, such as a gray letter or a confirmed square. Choice means the guess was compatible with the board but did little to distinguish the remaining candidates. Vocabulary means the answer was an unfamiliar word or spelling.
Those categories lead to different practice. An evidence problem calls for a checking routine. A choice problem calls for comparing candidates. A vocabulary problem calls for learning a word. Do not assign yourself twenty opening-word exercises because you missed a spelling you had never encountered. Also leave room for 'no clear mistake': a loss can happen even when your decisions were reasonable.
Describe the choice before judging it
Write down what you knew at the moment of the decision. If there were four candidates and two guesses left, record that small family. Looking at the revealed answer makes your losing guess seem much sillier than it may have been. The useful question is whether another available choice handled those four candidates better, without borrowing knowledge from the result screen.
For example, 'picked the most familiar word' is only a problem if you can show what that choice left unresolved. If the alternatives shared all but one square, the real issue may have been the size of the family and the number of rows remaining. You cannot repair that by promising to be less unlucky. You can inspect whether an earlier row could have split the family sooner.
Review after ten entries, not after every annoyance
At the end of ten recorded rounds, count the labels. Imagine six evidence entries, two vocabulary entries and two rounds with no obvious error. The next experiment should probably be an evidence check, even if an impressive new opener is more tempting. Choose one change and keep the rest of your routine steady for the next session.
A workable experiment is: 'Before submitting, compare the word against every green and every recorded yellow exclusion.' A vague goal such as 'concentrate harder' gives you nothing observable to review. The point of the next ten entries is to see whether the same mistake still appears, not to guarantee a particular solve average from a tiny sample.
Keep the notebook easy to abandon and restart
Use a note on your phone or a few lines on paper. Leave a blank template at the top: date, level, row, evidence available, choice made, replacement habit. If you miss a week, resume with the next interesting round. Leave forgotten rounds blank; a reconstructed memory is a shaky basis for the next experiment.
Retire a habit once it becomes reliable. You do not need to keep praising yourself for checking greens forever. Pick the next recurring problem, or stop recording for a while and enjoy the game. The log has succeeded when it changes a decision you can recognize; it does not need to become another daily obligation.