Where this goes wrong
Swapping the definitions. Trials of 97.9–102.1 °C averaging 100.0 °C get called "precise, since the average is right." An average on the true value is accuracy. Precision is agreement among the trials: relative range 4.2%, above 2% → not precise.
Judging the set by one trial. In 51.42, 51.38, 51.44, 51.40 g against a true 50.00 g, trial 2 sits 51.38 − 50.00 = 1.38 g from the true mass, yet the set is precise: relative range 0.12%, well under 2%. Both verdicts describe the whole set, never one reading.
Counting digits as accuracy. A caliper reads 14.42 cm, four digits, on a rod whose true length is 15.00 cm: the reading is 0.58 cm short. Digits show how finely the scale reads, not whether the reading is right.
Treating precise as accurate. Four readings within 0.06 g of one another all sit 1.41 g above the true mass: percent error 2.8%, above the bar. Agreement rules out scatter; it cannot rule out an error shared by every trial.