Password Strength Calculator

Type a password to see its entropy, strength rating and estimated brute-force crack time.

Result

65.7 bits
Good
190 years
95

How it is calculated

The engine starts from length and the kinds of characters used. The pool is the sum of the sets present: lowercase letters add 26, uppercase 26, digits 10, and any other character (symbols, spaces, accented letters) adds 33, the 32 ASCII punctuation marks plus the space, so all four classes give 95, the printable ASCII characters with the space (NIST's withdrawn SP 800-63-2 used 94, without it). Then entropy (bits) = length × log₂(pool size): that is the entropy the password would have if it had been chosen at random from that pool, which is the only case SP 800-63-2 gave the formula for; a password a person makes up has less, so read the figure as an upper bound. Two discounts follow. A password made of one short unit repeated (aaaa, abab) is only credited with the entropy of that unit. A well-known word such as password, qwerty or admin, even with digits or symbols added at the end, and plain number runs like 123456, is capped at 20 bits. Crack time is 2entropy ÷ 10¹⁰ seconds, assuming 10 billion guesses per second, shown in seconds, minutes, hours, days or years, and shown as "> 1,000,000,000,000,000 years" beyond that. The rating is Weak below 40 bits, Moderate from 40 to 60, Good from 60 to 80 and Strong from 80; these bands and the 10 billion guesses a second are the tool's own yardsticks, not a standard. For what NIST recommends now, see SP 800-63B-4: at least 15 characters for a password used on its own, a blocklist check and no composition rules; it gives no entropy figures. An empty field gives no result. The tool has no leaked-password list, so a predictable password outside its short built-in list can still score higher than it deserves.

Worked example

Take the password Maple7!Fox, which has 10 characters. It contains lowercase and uppercase letters, a digit and a symbol, so the pool is 26 + 26 + 10 + 33 = 95. The log₂ of 95 is about 6.57, and 10 × 6.57 gives 65.7 bits of entropy. It is not a repeated unit and not a listed common word, so no discount applies. That falls in the 60 to 80 band, so the rating is Good. At 10 billion guesses per second, exhausting all 265.7 combinations takes about 190 years. By comparison, the common word password is capped at 20.0 bits (Weak, under a second).

Try this example →

Common mistakes

  • Typing a password you actually use. The text you enter is sent in the page address (as ?password=…), so it stays in your browser history and can appear in server logs; test a similar-looking pattern, not the real thing.
  • Trusting a high score for a predictable choice. A name or phrase with a capital letter, a digit and a symbol, such as Tr0ub4dor&3, scores 72.3 bits and Good, although attackers try such patterns early. Only a short list of very common words is discounted.
  • Expecting a long repeated string to look strong. Twenty letter a characters score only 4.7 bits, because a repeated unit carries the entropy of the unit.
  • Reading "< 1 sec" or "9 min" as a bad calculation. Short passwords really do fall in that time under the 10 billion guesses per second assumption; a nine-letter lowercase password shows 9 min.
  • Reading the crack time as a guarantee. It assumes one fixed guess rate against one password; real attacks depend on how the password is stored and are far faster or slower.

Sources

About this calculator

A password's entropy measures how many bits of randomness it contains. For a password chosen at random, it is the length multiplied by log₂ of the size of the character set it is drawn from (the formula in the appendix of NIST's withdrawn SP 800-63-2, which gives it for randomly chosen passwords only); a password a person makes up carries less, so this tool's figure is an upper bound — a longer password, or one drawn from a larger pool of possible characters (lowercase, uppercase, digits, symbols), has higher entropy and more possible combinations an attacker would need to try. The estimated brute-force crack time then comes from dividing the total number of possible combinations (2 raised to the entropy in bits) by an assumed number of guesses per second. The 10 billion guesses a second and the 40, 60 and 80-bit bands are this tool's own yardsticks, not a standard.

Security teams, IT administrators and anyone designing a password policy use entropy and crack-time estimates to explain why ‘P@ssw0rd1’ is weak despite looking complex — its character set is guessable and it is short — while a longer passphrase built from unrelated words can have far higher entropy despite looking simpler, an approach reflected in modern password guidance: NIST's SP 800-63B-4 asks for at least 15 characters for a password used on its own (8 with a second factor), a check against a blocklist of common and breached passwords, and no composition rules, and it no longer gives entropy estimates.

Was this helpful?

You can pick up to 3 reactions.

Comments

  • Be the first to comment.

Popular calculators

All Calculators