SQLite: How to avoid using two CTE when using CASE statement?

I have a table trx with following schema:

| id   | p_id |
| 1    | 1    |
| 2    | 1    |
| 3    | 4    |
...
| 1000 | 2    |

where id is the transaction id and p_id the id of the person doing it.

I need to query trx so I get a table that allows me to plot an histogram of frequencies of transactions, that means, I want to know how many p_id did only 1 transaction, how many did 2, and so on. I have to aggregate 11 and more transactions in a +10 bin:

| n_trx | bin_size |
| 1     | 10       |
| 2     | 18       |
| 3     | 7        |
...
| +10   | 26       |

I know that I need the CASE statement for the +10 thing, and I approached the task using two CTE:

WITH new_trx_history
     AS (WITH trx_history
              AS (SELECT p_id,
                         Count(DISTINCT id) AS n_trx
                  FROM   trx
                  GROUP  BY p_id)
         SELECT CASE
                  WHEN n_trx < 11 THEN n_trx
                  ELSE '+10'
                END      AS n_trx,
                Count(*) AS bin_size
          FROM   trx_history
          GROUP  BY n_trx)
SELECT n_trx,
       Sum(bin_size)
FROM   new_trx_history
GROUP  BY n_trx; 

I would like to know if there is a way to have a more direct approach than my actual (working) query.

Here is the SQLFiddle

Answer

You can use CASE in GROUP BY

     WITH trx_history
          AS (SELECT p_id,
                     Count(DISTINCT id) AS n_trx
              FROM   trx
              GROUP  BY p_id)
     SELECT CASE
              WHEN n_trx < 11 THEN n_trx
              ELSE '+10'
            END      AS n_trx,
            Count(*) AS bin_size
      FROM   trx_history
      GROUP  BY CASE
              WHEN n_trx < 11 THEN n_trx
              ELSE '+10'
            END
      ;