Journal Citation Reports (JCR) and Essential Science Indicators (ESI) are two distinct but related products. Both products group journals into categories and rank journals (eg, quartiles) in those categories. But the precise set of categories between the two products will differ: JCR uses the 252 Web of Science Categories, ESI uses 22 categories specific to ESI. Consequently, each JCR category is much more specific and tailored to a narrow field, compared to ESI where each category represents a broad discipline. As a result, a single ESI category tends to comprise the journals from multiple JCR categories.
One of the facts of scholarly publication is that different research fields tend to have different publication and citation patterns. Perhaps earning 50 citations in category X represents a very high achievement (Q1). But in category Y, this is below average (Q3). If the journals from these two JCR categories get collapsed under a single ESI category Z, the journals that are Q1 in JCR category X will drop toward Q3 in ESI category Z. Conversely, the lower quartile journals of JCR category Y will tend to rise toward Q1 in ESI category Z.
In addition, JCR ranking is based on the Journal Impact Factor (JIF), which measures the ratio of citations vs source papers published in a journal whereas ESI ranks journals merely based on citation counts meaning that larger journals are likely to get a boost in ESI (by virtue of publishing many items that can earn citations) that they will not get in JCR (because the increase in source papers tends to drive down the JIF).
Both of these factors (category size and ranking method) create room for differences in ranking between the two products, even for the same journal.
(Clarivate Analytics)