For example, I have two logically equivalent constructions:

• Get 10 records with one query (avg single query time = 2.1 ms):

  getRows("SELECT * FROM foo WHERE id IN (1, 12, 23, 34, 45, 56, 67, 78, 89, 100)");
  

  QueryStatistics internally: totalExecutionTime = 2100, executionCount = 1000
  As result: getExecutionAvgTime() * getExecutionCount() = int(2100 / 1000) * 1000 = 2 * 1000 = 2000 ms

• Get 10 records with 10 queries (avg single query time = 0.9 ms):

  for (int id : new int[] {1, 12, 23, 34, 45, 56, 67, 78, 89, 100})
  {
    getRow("SELECT * FROM foo WHERE id = ?", id);
  }
  

  QueryStatistics internally: totalExecutionTime = 9000, executionCount = 10000
  As result: getExecutionAvgTime() * getExecutionCount() = int(9000 / 10000) * 10000 = 0 * 10000 = 0 ms

As a result, according to statistics slower version is a more rapid as performed instantaneously.