Those characters never show up in the text of the typing session. They were in the statistics only because they were typed mistakenly.
For example, if I typed "X" but the text doesn't contain the "X" at least once, then "X" would be counted as 0.00%-accuracy character, which doesn't make sense as the accuracy is meant for those characters supposed to be typed instead of those mistakenly done.
Those characters never show up in the text of the typing session. They were in the statistics only because they were typed mistakenly.
For example, if I typed "X" but the text doesn't contain the "X" at least once, then "X" would be counted as 0.00%-accuracy character, which doesn't make sense as the accuracy is meant for those characters supposed to be typed instead of those mistakenly done.