What Does "99.99% Probability of Paternity" Actually Mean?

The two possible result categories

Every DNA paternity test produces one of two results:

• Exclusion: The alleged father is not the biological father. This result is reported at 0 percent probability of paternity. There is no statistical wiggle. The alleged father lacks the genetic markers required to be the biological parent, and that is a definitive negative.
• Inclusion: The alleged father is the biological father. This result is reported at 99 percent or higher probability of paternity. Specifically, 99.99 percent is the typical reported result, though some tests return 99.999 percent or 99.9999 percent depending on the genetic markers.

Notice the asymmetry. Exclusion is 0 percent. Inclusion is 99.99 percent. There is no symmetric 100 percent for inclusion. Why not?

What 0 percent exclusion really means

A child inherits exactly half of their DNA from each biological parent. At every genetic marker tested (typically 16 to 24 markers), the child has two variants: one from the mother, one from the father.

If the alleged father does not have one of the variants the child carries at multiple markers, he cannot be the biological father. There is no statistical way around this. He simply does not have the DNA that the biological father would have to have. The exclusion is definitive at every level.

This is why DNA paternity testing is sometimes described as more reliable at excluding a man than at confirming him. Exclusion is logical certainty. Inclusion is statistical confidence.

What 99.99 percent inclusion really means

An inclusion result means: at every marker tested, the alleged father has at least one of the variants the child carries. He matches the genetic profile required to be the biological father.

The 99.99 percent number is a probability calculation. Specifically, it represents the answer to this question: "Given that the alleged father matches the child at every tested marker, how likely is it that he is the actual biological father rather than a random unrelated man who happens to match by coincidence?"

The math accounts for:

• How common each matching genetic variant is in the general population
• How many markers were tested (more markers, higher confidence)
• Whether the mother's DNA was included (trio testing is slightly higher confidence than duo)
• The ethnic background reference frequencies used by the lab

A 99.99 percent result means there is roughly a 1-in-10,000 chance that an unrelated random man would coincidentally match the child at every tested marker. For a 99.999 percent result, the odds drop to roughly 1-in-100,000. For 99.9999 percent, 1-in-1,000,000.

Why not 100 percent

The theoretical reason it is not reported as 100 percent: there is always some non-zero probability that an unrelated random person happens to share all the same genetic markers tested by pure chance. The probability gets vanishingly small as more markers are tested, but it never reaches mathematical zero.

The practical reason: there is also the theoretical possibility of a close biological relative (like an identical twin brother of the actual father) matching all markers. Most labs do not assume this possibility, but the reporting convention preserves the statistical honesty by stopping just short of 100 percent.

Functionally, a 99.99 percent result is treated as definitive paternity by every court and government agency that accepts DNA testing. The remaining mathematical uncertainty is not material to legal or practical purposes.

The Combined Paternity Index (CPI)

Most DNA paternity test reports include something called the Combined Paternity Index, often abbreviated CPI. This is a number, sometimes very large (like 100,000 or 1,000,000), that represents the strength of the genetic evidence.

The CPI is the underlying calculation that the probability percentage is derived from. It compares two scenarios:

• How likely the observed genetic results are if the alleged father is the biological father
• How likely the observed genetic results are if a random unrelated man is the biological father

A CPI of 100,000 means the alleged father is 100,000 times more likely to be the biological father than a random unrelated man. The probability percentage is calculated from the CPI using standard statistical formulas.

You do not need to interpret the CPI yourself. The probability percentage is the takeaway for almost all practical purposes. But if you see a CPI on your report, that is what it means.

What if the result is between 99 and 99.99 percent?

Sometimes a paternity test result comes back at 99.5 percent or 99.7 percent rather than 99.99 percent. This usually happens when:

• The mother was not tested (motherless tests are typically reported slightly lower)
• The genetic markers happen to align in a less statistically powerful way
• The alleged father and the child have unusual or rare variants at some markers
• A close biological relative may also be a possible father (which the test cannot distinguish without additional testing)

Any result of 99 percent or higher is considered legally sufficient for paternity in California. The difference between 99.5 and 99.99 is real but not practically meaningful in court. Both establish paternity for legal purposes.

What courts actually require

California Family Code section 7555 sets the threshold for presumed paternity at 99 percent. Most labs return results well above this threshold (99.99 percent is the typical inclusion result for a standard trio test), so the threshold is rarely a close call.

For federal purposes such as Social Security, the same general threshold applies: 99 percent or higher probability of paternity for an inclusion result.

If a test returns below 99 percent, it is considered inconclusive rather than positive. The next step is usually to add the mother (if not already included), test additional markers, or test another close relative for confirmation.