Why does a completed crawl show a different number of pages crawled versus when it was running?

While a crawl is running, the Site Health report will display every single URL that has been encountered during the crawl so far. Once the crawl is completed, the Audit reports are processed and duplicates are removed which can lead to a difference in the number of URLs crawled from the time the crawl was running.

An example of when this occurs is If multiple URLs redirect to the same URL, that URL is counted multiple times in the count of pages crawled.