How to use the list batch splitter
- Paste your list, one item per line. Blank lines are skipped.
- Choose Batches of a size or A number of groups, and enter the size or the number.
- For groups, choose whether to deal the items out like cards (item 1 to group 1, item 2 to group 2, and so on) or keep neighbors together.
- Optionally shuffle the list first with a seed number, then copy the groups or download a CSV file with one row per item.
Formula
batches of n: group k holds items (k-1) x n + 1 to k x n; the last group holds the rest
k groups in blocks: the first (items mod k) groups hold one extra item
k groups dealt: item i goes to group (i - 1) mod k + 1
In both ways of making k groups the sizes differ by at most one. Shuffling uses a seeded generator and the Fisher-Yates method, so the same list and seed give the same order every time.
Twenty-three items
The 23 items Item 1 to Item 23 in batches of 10 make three groups of 10, 10 and 3. Split into 4 groups in blocks, the sizes are 6, 6, 6 and 5, because 23 = 4 x 5 + 3 and the first three groups take one extra.
Dealt into 4 groups, group 1 gets Items 1, 5, 9, 13, 17 and 21, and the groups still have 6, 6, 6 and 5 items. The CSV download has 23 rows plus a header.
Blocks or dealing
Blocks keep an ordered list in runs, which suits a list sorted by date or by name where neighbors belong together. Dealing mixes the order, which suits sharing tasks fairly or making teams that should not be neighbors in the list. Shuffling before either splits the list at random.
The shuffle is for fair looking splits and repeatable results, not for security or lotteries. For a secure shuffle see the List Randomizer, and to build teams balanced by skill use the Balanced Team Generator. To cut a long text into chunks instead, see the Text Chunk Splitter.
Using the output
Copy result gives every group under a heading with its size. The CSV has two columns, the group number and the item, which opens in any spreadsheet. A split into more than 500 groups is refused, since the table would be too long to read.
Assumptions and limits
- Each non-blank line is one item, trimmed of spaces at both ends, and repeated items are kept.
- Up to 20,000 lines and 500 groups are accepted.
- The seed is a whole number from 0 to 4,294,967,295 and is used only when the shuffle box is ticked.
- If there are more groups than items, the extra groups are empty and shown that way.
- The tests compare the shuffle and every split with a separate Python implementation.
Frequently asked questions
Why do I get the same shuffle every time?
The seed fixes the order. Change the seed number to get a different shuffle.
Which is fairer, blocks or dealing?
Both keep sizes within one item. Dealing spreads items from every part of the list across the groups, so it is the better choice for the end of a sorted list.
Can I keep the order inside each group?
Yes. Without shuffling, items stay in their original order inside each group.
What happens to a short last batch?
It holds whatever is left, so batches of 10 from 23 items end with a group of 3.