The index selection block prints out an estimate of the number of rows that would be returned if this index were used and the number of pages that would need to be read. It includes the index height.
For a single-table query, this information is basically all that is needed for the optimizer to choose between a table scan and the available indexes. For joins, this information is used later in optimization to help determine the cost of various join orders.
The three cluster ratio values for the index are printed, since estimates for the number of pages depend on cluster ratios.
If the index covers the query, this block includes the line:
Index covers query.
This message indicates that the data pages of the table do not have to be accessed if this index is chosen.
