Password Entropy Explained: Bits, Length, and How Long a Password Takes to Crack

What a bit of entropy actually means

Entropy counts how many equally likely possibilities a password could have been, expressed on a logarithmic scale. One bit means two possibilities, two bits means four, ten bits means 1,024. Because the scale is logarithmic, adding a single bit doubles the search space an attacker must sweep through. A password worth 40 bits sits somewhere among about a trillion candidates, while 60 bits reaches a quintillion.

The formula only holds when the password is chosen at random. If a human picks it, real entropy collapses because people favor names, dates, and predictable substitutions like a zero for the letter O. A tool estimates entropy from the character set and length, but that number is an upper bound: it assumes every position was rolled independently, which is exactly what a password manager does and a tired human does not.

Length beats complexity

Adding a character type widens the alphabet, but only once. Moving from lowercase letters (26 symbols) to lowercase plus digits (36) to a mixed set of 94 printable symbols raises the per-character entropy from about 4.7 bits to 5.2 to 6.6. Adding one more character, by contrast, adds those bits again for every extra position. Twelve random lowercase letters carry more entropy than eight characters from the full symbol set.

This is why passphrases work. Four random words from a 7,776-word list carry about 51 bits, roughly the strength of a 10-character symbol soup, yet they are far easier to type and remember. The lesson is to lengthen first and only widen the alphabet after length is exhausted, since length compounds while alphabet changes do not.

From bits to crack time

To turn entropy into time, divide the number of possibilities by an attacker's guess rate. A mid-range GPU rig can try billions of guesses per second against a fast hash, so a password with 44 bits, around 17 trillion combinations, falls in hours. Add 20 bits to reach 64 and the same rig needs thousands of years, because those 20 bits multiply the work by roughly a million. This is why small increases in length produce dramatic jumps in safety.

Crack time also depends on how the password is stored. A slow, salted hash such as bcrypt or Argon2 deliberately caps the guess rate to thousands per second, buying enormous margin. A leaked database that used a fast unsalted hash gives no such protection, so the same password that looks safe in one system is trivial in another. Treat crack-time estimates as offline worst cases and pick a length that survives them.

Frequently asked questions

How many bits of entropy is enough?

For accounts protected by a slow hash and rate limiting, 60 to 70 bits is comfortable. For high-value secrets or fast-hash exposure, aim for 80 bits or more, which usually means a long random passphrase or a manager-generated string.

Does adding symbols really help less than length?

Yes. Widening the alphabet raises entropy per character once, while each extra character adds entropy again and again. Beyond a mixed set, one more character usually beats one more symbol type.

Is a password reused across sites still safe if it has high entropy?

No. Entropy protects against guessing, not against reuse. If one site leaks the password in plain text or a weak hash, attackers try it everywhere. Use a unique high-entropy password per site.

Why does a password checker rate my clever password as weak?

Good checkers model human habits, so they discount dictionary words, keyboard patterns, and predictable leetspeak substitutions. Your password may be long, but if its structure is guessable the effective entropy is far below the theoretical maximum.