IDs are assigned by crt.sh in ingestion order, and so can be heavily biased depending on which log was ingested at what time and which CAs logged there. I don't know what the error is, but relying on them for anything date-based seems unreliable.
I don't rely on them for anything date-based, with one exception. I use them as a hint for the time of issuance, to obtain a statistical sample, because there is high correlation between ID and date of issuance. In the end, I use the date of issuance of the certificates I actually download. The one exception is the estimated number of leaf certs per quarter, but that's a statistic that I'm less interested in... other than this dip in the last few quarters.
but you need to be smart about how you structure your query to be re-entrant as the postgres API doesn't support long-running queries.
I think there are too many complications with the database for me to construct a meaningful query... although perhaps someone else may be interested in the same questions I am, and may take this investigation further. Anyway I have enough data of the type I wanted to make reasonable conclusions.
