Strictly speaking the MAX
types will always be a bit slower than the non-MAX types, see Performance comparison of varchar(max) vs. varchar(N). But this difference is never visible in practice, where it just becomes noise in the overall performance driven by IO.
Your main concern should not be performance of MAX vs. non-MAX. You should be concerned with the question it will be possible that this column will have to store more than 8000 bytes? If the answer is yes, even by if is a very very unlikely yes, then the answer is obvious: use a MAX type, the pain to convert this column later to a MAX type is not worth the minor performance benefit of non-MAX types.
Other concerns (possibility to index that column, unavailability of ONLINE index operations for tables with MAX columns) were already addressed by Denis' answer.
BTW, the information about the columns over 4KB having remaining data in an overflow area is wrong. The correct information is in Table and Index Organization:
ROW_OVERFLOW_DATA Allocation Unit
For every partition used by a table
(heap or clustered table), index, or
indexed view, there is one
ROW_OVERFLOW_DATA allocation unit.
This allocation unit contains zero (0)
pages until a data row with variable
length columns (varchar, nvarchar,
varbinary, or sql_variant) in the
IN_ROW_DATA allocation unit exceeds
the 8 KB row size limit. When the size
limitation is reached, SQL Server
moves the column with the largest
width from that row to a page in the
ROW_OVERFLOW_DATA allocation unit. A
24-byte pointer to this off-row data
is maintained on the original page.
So is not columns over 4KB, is rows that don't fit in the free space on the page, and is not the 'remaining', is the entire column.
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…