Two people are injured in identical accidents, with identical injuries, and identical damages of 100,000. Both are found 50 percent at fault. One recovers 50,000. The other recovers nothing. The difference is not the facts of the accident; it is which negligence rule applies.
There are four rules in use, and the gap between them is largest exactly where most disputed cases land.
Pure comparative
Recovery is reduced by the plaintiff's share of fault, and that is all. There is no cut-off.
At 20 percent fault on a 100,000 award, recovery is 80,000. At 75 percent fault it is 25,000. Even a plaintiff who is overwhelmingly responsible recovers something, because the only operation is the reduction.
Modified, barred at 50 percent
The same reduction applies, but recovery is barred once the plaintiff's fault reaches 50 percent.
At 49 percent fault, recovery is 51,000. At 50 percent, recovery is zero. One percentage point of fault moves the outcome by 51,000 on these numbers, and that is the entire award.
Modified, barred above 50 percent
Identical, except the bar applies only once fault exceeds 50 percent. Fifty percent exactly still recovers.
At 50 percent fault, recovery is 50,000. At 51 percent, zero.
Compare that with the previous rule. Both are called "modified comparative" and both bar recovery around the halfway point, but at exactly 50 percent fault one pays 50,000 and the other pays nothing. Same facts, same finding, opposite result.
Contributory negligence
Any fault at all bars recovery entirely. At 1 percent fault the recovery is zero, and the same is true at 10 or 20 percent.
This is the harshest rule by a wide margin, and it is why the apportionment of fault is fought over so hard in jurisdictions that use it. Under pure comparative, moving a plaintiff from 10 to 20 percent fault costs them 10,000 on a 100,000 claim. Under contributory negligence, moving them from 0 to 1 percent costs them everything.
Where the rules diverge most
Run the same 100,000 claim through all four at a few fault levels and the pattern is clear.
At zero fault every rule pays 100,000; the rules are indistinguishable. At 10 or 20 percent, three of the four pay identically (90,000 and 80,000) and only contributory pays nothing. The rules only separate from each other near the halfway mark, which is precisely where contested liability tends to be argued.
That is the practical point. If fault is clearly minimal or clearly overwhelming, the choice of rule often does not change the number much. If fault is genuinely disputed and lands near half, the rule can be worth the entire claim.
What this does not tell you
Which rule applies to a given case is a question of the law where the claim is brought, and that is not something the arithmetic can answer. The four models above describe how each rule behaves once you know which one governs. Determining that is a separate question and a legal one.
For the arithmetic across all four rules at once, WorthMyClaim on one percentage point fault works the numbers through side by side.