September 23, 2026
Why "What's My Case Worth?" Is the Wrong Question
Ask what a case is worth and you get a single number. That number hides the two things that actually decide the outcome: how often a case like this pays at all, and how widely the payouts spread. Here is the better question, and why the answer you want is a range.
Category: Legal Insights|Reading time: ~4 min
It is the first question a client asks and the first one a lawyer answers: what is the case worth? It feels like the right question. It is the wrong one, because it asks for a single number when the honest answer is a distribution.
A case does not have a worth the way a house has a price. It has a range of outcomes, each with its own likelihood. Collapse that into one figure and you throw away the two things that matter most: whether the case pays at all, and how far the outcomes spread.
A single number is an average of outcomes that never happen
When someone says a case is "worth $400,000," they usually mean the average of comparable outcomes. But no individual case pays the average. Some pay nothing. Some pay far more. The average is a summary of a crowd, and your case is one person in it.
Two numbers are buried inside that average, and both change the decision:
- How often a case like this pays anything. A profile that pays one time in five is a different proposition from one that pays three times in five, even if the payouts are identical when they land.
- How widely the payouts spread. Two profiles can share the same average and look nothing alike: one clusters tightly around the middle, the other has a long tail of large results.
Ask only "what is it worth?" and you see neither. You get a point where you needed a shape.
The jurisdiction problem, in one comparison
If a case had a single worth, it would not change when you crossed a state line. It does.
A comparable case with a median payout around $73,000 in California carries a median near $507,000 in Pennsylvania. Same injury, same allegation, roughly seven times the figure. "What is it worth?" has no answer until you say where it is filed. The question assumes a fixed value that does not exist.
Severity does the same thing, and not in the direction most people expect. Cases that end in death carry a lower median payout than cases that end in grave permanent injury. A single "worth" hides that inversion completely.
The better questions
Replace the one question with two:
- How likely is any payment? This is the threshold. It tells you whether the case is worth pursuing before you argue about the amount.
- If it pays, how wide is the range? This tells you how much is riding on the tail, which is what should drive settlement posture and risk.
Answer those two and the single number reassembles itself into something useful: a probability, a realistic range, and a sense of how volatile the outcome is. That is what a prediction from the Hooper Engine actually produces, and it is why the output is a range and a percentage rather than a lone figure.
What the right question changes
Framing the question as a distribution changes real decisions:
- At intake, you weigh a low-probability, high-payout case against a high-probability, modest one on the same honest scale, instead of chasing the biggest headline number.
- At mediation, you argue from the spread. If a meaningful share of comparable cases resolved above the number on the table, that is leverage a single average cannot give you.
- With a client, you can be honest. "Cases like yours pay about 30 percent of the time, and when they do, most land between here and here" is more useful, and more trustworthy, than a confident figure you cannot actually stand behind.
The honest limit
A distribution is still not a guarantee. Your specific case will land somewhere in the range based on facts, witnesses, and judgment that no dataset captures. The distribution is the base rate, not the verdict.
But a base rate is exactly what "what is it worth?" pretends to give you and cannot. The question sounds precise and delivers false precision. The better questions sound less certain and deliver something you can actually use.
Try It
Next time you would ask what a case is worth, run it instead and read the probability and the range. The answer is less tidy and far more honest.
MedMalPredict AI is not legal advice. Predictions are based on historical data and represent probabilities, not guarantees.
Enjoyed this article?
Subscribe to get insights on medical malpractice trends, AI in litigation, and data-backed case analysis. No spam, unsubscribe anytime.