Although there is nothing special about this query from the explain results, the following points can be considered.
1, shippingtype and status appear to be enumeration values ​​and can be replaced by tinyint (numeric query is faster than string query);
2, endtime are set to NOT NULL , replace the original null value with a special value (-1 or 0). The NULL value will affect the efficiency of the index;
3. If possible, use user_id instead of user;

Optimization of personal YY:

SELECT count(*) FROM trade 
WHERE status IN ("TRADE_FINISHED", "WAIT_SELLER_SEND_GOODS") 
AND shippingtype <> "free"
AND tosellerreachgoods = 0 
AND user = "xxxxxxxxxxxxxxxx"
AND consigntime < 1470110400000 
AND created BETWEEN 1469980800000 AND 1470067200000
AND ( endtime >= 1469980800000 OR endtime IS NULL );

There is nothing much to say about the first four conditions. The main ones are the last three conditions, which are consigntime, created and endtime. They are arranged according to the filtering range from small to large (specifically, it depends on the three conditions in your table. range of fields).

Some suggestions are based on the fact that you can modify the table structure. If you don't have permission, just ignore it.

Reference: http://tech.meituan.com/mysql...

Strong

That does not mean affecting performance

From the explanation, there is no problem