The biggest ending on this site is the one with no name

The ending filter offers five: -a, -o, -y and -ie, -i, -s. Each has a page on this site.

What it doesn't offer is the sixth option, which is everything else, and everything else is larger than any of the five: 4,848 names against 2,312 ending in -a and 1,874 in -y or -ie.

There is a sixth ending in there

Sort the leftover by its last letter and the largest group isn't a scatter. It's -e: around eleven hundred names, more than a fifth of the bucket.

Blue, Rose, Bruce, Jade, Sage, Clementine. A silent -e behaves nothing like a hard consonant at the end of a name — it lengthens the vowel before it and leaves the name open — which is the property the other five endings were named for.

So the filter's five named endings are five, and the sixth is sitting inside the leftover unnamed. That isn't a data error; the classification simply stopped where somebody stopped. It's worth knowing before you read the bucket as a single thing.

The rest of it is genuinely mixed

Past -e the last letters run n, r, t, l, u, h, k — and those don't behave alike either.

1,389 of the bucket classifies strong-sounding and 601 soft, which is the widest spread of any ending group here, because ending in a consonant says nothing about which consonant.

It leans male — 2,033 measure male against 919 female — though not the hardest: -o and -s both lean further. A closed ending reads male in this catalogue wherever it appears, and this is where most of the closed endings live.

The consonants that actually differ

If you're going to work in this bucket, it's worth knowing roughly how the common endings behave, because the filter won't tell you and the difference is audible.

Stops — t, k, p, d, b, g — end the name dead. Pip, Bud, Jet, Hawk. They're the sharpest thing in the catalogue, they carry a short distance very clearly, and they're the ones most likely to collide with a training cue, because cues are built the same way.

Nasals — n, m — can be held. Finn, Sam, Bean, Storm. You can lean on the last sound and stretch it, which gives a closed ending some of the carrying power of an open one. This is the most underrated corner of the bucket for a dog that's called across distance.

Liquids — l, r — are soft closings that trail instead of stopping. Pearl, Bear, Hazel, River. They read warm, they don't cut through noise particularly well, and they almost never clash with a cue.

Choosing between those three is a more useful decision than choosing between the named endings, and nothing on the site sorts by it. The nearest available proxy is the sound filter, which splits this bucket more evenly than it splits any other.

Length is the part that is uniform

2,644 of it is two syllables and 1,606 a single one, with only 598 running to three or more.

Consonant endings are short endings: there's no equivalent of the three-syllable -a name that gets clipped in the second week.

They average 63 registrations each, in the middle of the endings — under -y and -ie at 129, above -a at 53. Nothing about ending in a consonant costs or earns use.

That stability is the bucket's quiet advantage. Because the shape doesn't drive popularity, the crowding here is entirely about the individual name, which means you can pick on sound and check the company separately instead of trading one against the other.

The hard ending and the training cue

The one genuine practical problem in this bucket is that a closed ending is the same shape as a command.

Sit, stay, down, come, heel, no, off, wait — every one of them ends in a consonant, and most of them are one syllable. A one-syllable name with a hard ending is being asked to sit next to eight words built exactly like it, and the animal separates them by vowel long before it separates them by consonant.

So the test matters more here than anywhere else on the site, and it takes two minutes. Write the cues you actually intend to teach, then say each shortlist name immediately after each one, out loud. Anything that sounds like an echo comes off the list.

The commonest collisions are the obvious ones once heard: a name with the vowel of sit, one with the vowel of no, one that rhymes with stay. And the commonest fix is a second beat — a two-syllable name with the same closed ending keeps the edge and stops sounding like an instruction.

None of this argues against the bucket. 2,644 of it has two syllables already, which is where the problem mostly solves itself.

Why this is where the word-names live

English nouns mostly end in consonants, and that single fact explains most of what's in this bucket.

If you want a name that's an ordinary word — a colour, a plant, a tool, a place, a food, a weather — you'll almost certainly end up here, because that's where the vocabulary of the language sits. The named endings are largely populated by given names and diminutives; this one is populated by things.

1,322 of the bucket are names a dictionary lists for people, which is a lower proportion than the human-name endings carry and is exactly what you'd expect from a group that's mostly words.

It's also why this is the most reliable place to find a name nobody else has. A word chosen well needs no explanation, carries a meaning you picked deliberately, and hasn't been through a century of use as a human given name — which is the main reason the other slices are crowded.

A note on the counting itself

The bucket exists because a classification had to stop somewhere, and it's worth being clear about what that means for the numbers on this page.

Every figure here describes a group defined negatively — the names that aren't one of five things — so it's an average over a set that has no shared property beyond exclusion. That's why the sound split is the widest in the catalogue and why the average registration count sits so neatly in the middle: you're measuring a remainder, and a remainder tends to look like the whole.

Treat the totals as real and the averages as weak. 4,848 names genuinely fall outside the five named endings; that number is a fact about the spelling. Whether they behave alike is a different question, and the answer is mostly no.

What to do with this

If a closed ending is what you want, the ending filter will get you there and the sound filter will sort the result: strong for names that arrive hard, soft for the ones that don't.

If it was the open -e you were after, no filter reaches it directly — read the list for the silent ones, or filter for soft-sounding and accept the overlap.

And if you're choosing for outdoor use, the practical order is: pick a nasal or a stop, keep it to one or two beats, then say it against your training cues. That combination is where the best-built called names in the catalogue come from, and it's three decisions rather than a search.

Then read the bucket as what it is. Four of the five named endings are one letter; the sixth is a rule about what a name is not. Those are different kinds of thing, and only one of them tells you how the name sounds.