Randomness only feels trustworthy if it's actually unbiased, so this picker uses the Fisher-Yates shuffle, the standard algorithm computer scientists use for shuffling a list fairly. In plain terms, it walks through your list from the last item to the first, and at each step swaps that item with a randomly chosen item from the remaining unshuffled portion. Every possible ordering of your list ends up equally likely, which means no name is secretly favored because of where it started in the list.
In "Pick One" mode, the tool shuffles your list and selects the first entry, while a quick visual cycle flashes through a few random names before landing on the winner, just for fun. The underlying selection is already decided by the shuffle, the animation is purely cosmetic. This is handy for choosing who answers a question next, who presents first, or who wins a small classroom raffle without anyone feeling singled out unfairly.
In "Split into Groups" mode, the tool shuffles the full list first and then divides it either into groups of a fixed size or into a fixed number of groups, whichever you choose. Shuffling before splitting is what keeps the assignment fair, if you simply split an unshuffled list, students who happened to be typed next to each other would always land in the same group. When the list doesn't divide evenly, the extra names are spread across the first few groups one at a time rather than dumped into a single oversized group.
Paste the full, current list every time rather than reusing an old one, stale names can quietly skew results if someone dropped the class. If you want to guarantee everyone gets picked eventually across a semester, keep a running list of who has already been chosen and remove them from future rounds. For groups, decide ahead of time whether uneven group sizes are acceptable, and if not, pick a group count that divides your class size evenly before shuffling.