Generate an ordered sequence of numbers with a chosen start, step and length, then shuffle it or keep it in order. Unlike independent draws, a sequence covers its range evenly — useful for sampling frames, test identifiers and lesson numbering.
Settings
Recent results
Generated locally in your browser — your settings and results never leave this page.
Sequences and draws answer different questions
An independent draw asks “give me a number”. A sequence asks “give me a set of numbers that covers this range”. The distinction matters more than it first appears, because the two produce very different-looking results even when they span the same interval.
Ten independent draws between 1 and 100 will cluster and leave gaps. That is not a fault — it is what independence looks like, and people consistently underestimate how uneven genuine randomness appears. A sequence has no gaps by construction: each value is the last plus a fixed step, so coverage is guaranteed.
Systematic sampling
The combination of a sequence and a shuffle is the basis of systematic sampling, a technique used wherever you need to inspect a representative slice of something too large to check entirely. Take every tenth record, every twentieth transaction, every fiftieth product from the line.
The advantage over picking at random is coverage: a systematic sample cannot accidentally take everything from the first half of the day. The risk is periodicity — if the underlying data has a repeating pattern that happens to line up with your step, the sample will be badly skewed. Sampling every seventh day always lands on the same weekday, which is exactly the trap.
Shuffling the order in which you work through the sequence does not remove that risk, but it does remove the predictability of when each inspection happens, which matters when the thing being inspected might otherwise be prepared for it.
Steps, direction and off-by-one
A positive step counts up and a negative one counts down. The count is the number of values produced, not the distance covered — a common source of confusion. Starting at 1 with a step of 10 and a count of 10 ends at 91, not 100, because the first value is the start itself.
This is the same off-by-one that makes fence-post problems awkward: ten posts have nine gaps between them. If you need the sequence to land on a specific final value, work backwards from it rather than assuming the count and the range line up.
Where sequences are the better tool
Test identifiers benefit from being sequential and readable — record 7 failing is easier to trace than a random eight-digit identifier failing. Lesson and exercise numbering wants even spacing. Load testing often wants a predictable spread of values rather than a clustered one, so that the same range is exercised on every run.
Anything that must be unguessable is the wrong use. A sequence is entirely predictable once you know the start and the step, so it should never be used for tokens, passwords or anything where the next value should not be inferable from the last.
Reproducibility
In seeded mode the shuffle becomes deterministic: the same seed and settings produce the same order every time. That turns a shuffled sequence into something you can put in a test suite, where a failure that cannot be reproduced is a failure that cannot be fixed.
Privacy
Sequences are generated in your browser. Nothing is transmitted, stored or logged.
Choosing a start and a step
The three settings interact in ways worth thinking about before generating. The start fixes where the coverage begins, the step fixes how wide the gaps are, and the count fixes how far the sequence reaches. Change any one and the end point moves.
For sampling, work from the population size. To take fifty items from five thousand, a step of one hundred spreads the sample evenly across the whole set; a step of ten would cover only the first five hundred and miss everything after. Dividing the population by the sample size gives the step that covers it exactly.
For numbering, the step is usually one and the start is whatever your system counts from. The classic mistake there is starting at one when the target counts from zero, which shifts every value by a position.
How to use the Random Number Sequence Generator
- Set the starting number, the step between values and how many values you want.
- Choose whether to keep the sequence in order or shuffle it.
- Generate, then copy the list or export it as CSV.
Frequently asked questions
How is a sequence different from just generating random numbers?
A sequence covers its range evenly. Ten independent draws from 1 to 100 might give you three numbers in the twenties and none above eighty; a sequence of ten starting at 1 with a step of 10 gives you one from each band. Use independent draws when each value should be unrelated to the others, and a sequence when you want even coverage.
What is the step for?
It is the gap between consecutive values. A step of 1 gives 1, 2, 3; a step of 5 gives 5, 10, 15. Steps can be negative to count down, which is useful for reverse numbering. A step of zero would repeat the same number forever, so it is not allowed.
Why would I shuffle a sequence?
Because it gives you even coverage in an unpredictable order — the basis of systematic sampling. If you need to audit every tenth record but do not want the order to be guessable, generate the sequence and then shuffle it.
Can I use this for test identifiers?
Yes, and it is a good fit. Sequential identifiers make it obvious which record is which while debugging, and shuffling the order they are created in exposes code that accidentally depends on insertion order.
Does the sequence ever repeat a value?
Not unless the step is zero, which is refused. Because each value is the previous one plus a fixed step, every entry in the sequence is distinct by construction — no duplicate checking is needed.
What is the largest sequence I can generate?
The count is capped so the page stays responsive and the output remains something you can actually read or paste. For genuinely large sequences, generating them in the system that will consume them is more practical than moving them through a browser.
Is the shuffle genuinely random?
It is a Fisher–Yates shuffle driven by your browser's cryptographic random source, which means every possible ordering is equally likely. Seeded mode is available when you want the same shuffle again from the same seed.
