Cheat Sheet: MySQL Explain

Syntax
EXPLAIN <SELECT STATEMENT>

Columns in Result
id	Statement Identifier
select_type	Which type of SELECT statement was given.
table	The name (or short handle) of the table in observation.
type	Describes how the MySQL Optimizer fits this table into the execution plan.
possible_keys	Lists the names of the indexes MySQL could use to optimize this particular statement.
key	Names the index that was actually used for this table in the given statement.
key_len	Length of the key in bytes.
ref	Which fields of the index are used (usefull for composite indexes).
rows	Estimation on how many rows will possibly be returned.
Extra	Useful hints and additions how the query will be executed.

Column: select_type
SIMPLE	A query which uses only one table or joins tables.
PRIMARY	Outermost SELECT Statement in query using UNION or Subselect.
UNION	Any second or later table in a UNION statement.
DEPENDENT UNION	Any second or later table in a UNION depending on an outer SELECT.
UNION RESULT	Resulting derived table of a union.
SUBQUERY	SELECT Statement within a subquery
DEPENDENT SUBQUERY	SELECT Statement within a subquery that is depending on an outer SELECT.
DERIVED	Indicates that the FROM Clause is a subquery.

Column: type
system	The table has only one value.
const	Evaluating a condition on a primary or unique key which leads to only one matching row.
eq_ref	For each combination of the other tables a row is read. Best possible join type fully utilizing primary or unique structures.
ref	For each combination of other tables all rows are read. Used when index in join are not unique.
ref_or_null	Like ref doing another pass to check for NULL values.
fulltext	Join uses a fulltext index.
index_merge	Indication that indexes are merged to join. Key field holds more than one index.
unique_subquery	Optimization in an subquery with an IN keyword having unique values.
index_subquery	Same as unique_subquery except that index values are non unique.
range	Statements with comparrision operators: LIKE, >=, >, <=, < IN, BETWEEN.
index	Sequential scan of index entries (includes covering index SELECT statements).
all	Indicates this table is accessed using a Full-Table Scan. This is the worst-possible outcome.

Column: Extra
Distinct	Searching for distinct values, so search can be stopped after first matching row.
range checked for each record	More than indexes are found but MySQL can't decide which is the best and tries.
Select tables optimized away	Aggregation functions used on a single table with no group by. Only matching one result.
Using filesort	Another pass is needed for sorting the result (no information if in memory or on table).
Using index	All operations are done using the index. Accessing the data was not needed.
Using temporary	MySQL needs to create a temporary table to resolve the query.
Not exists	LEFT JOIN optimization where only one pass is needed to match the criteria.
Using where	A where clause is used to restrict the rows accessed.