When, in United States v. Philadelphia National Bank (PNB), the Supreme Court blocked two Philadelphia banks from merging, banking was simple to measure. The product was checking accounts and loans, and the market was one city. The Court set out an arbitrary rule of thumb, deciding that if a merger gives you about 30% of a market, courts will assume it hurts competition. This standard was derived from the static and predictable market of bricks and mortar banking in a local area. But, by making up that 30% threshold, the case created the machinery to answer the question Congress wanted to address – whether a merger would substantially lessen competition. If PNB’s 30% market-share threshold is triggered by the states’ narrow market definition, however, it would create a legal presumption of harm the states are counting on to win, even without proving actual consumer damage.

There is a broad debate over the applicability of PNB to modern anti-trust cases, with many legal scholars finding the application of a random number, without any real analysis of the potential harm a merger might cause, to be a fatal flaw. In the Paramount case, the problem is more sharply defined because of how the states are using PNB. The rule is broadly triggered at the 30% threshold, and the states have drawn their market lines to get the number they need.

The math was easier in 1963 because banking in Philadelphia really was a closed world. If you wanted a checking account, you went to a local bank, period.

The states measure the Paramount deal the same way, as if entertainment still came in closed boxes. Their case counts wide-release theatrical movies and the basic cable bundle, and leaves out streaming, YouTube, sports rights, and everything else that now competes for the same hours of your evening. It is like declaring someone the tallest person in the room after sending everyone taller out the door.

Ask any household how they actually watch. YouTube is now the most-watched form of television in America. Streaming makes up nearly the majority of all TV time. Netflix and Amazon are buying up live sports. The complaint implies the combined company is dominant in a market that viewers are walking away from. The fact that plaintiffs can gerrymander market definitions to trigger PNB’s structural presumption is exactly the kind of abuse that has led scholars to argue PNB should be overturned. Here is the strange part. Under certain interpretations of the 1963 rule, courts are not allowed to weigh whether a merger might make competition stronger overall. The Supreme Court once said benefits in one area cannot excuse concentration in another, and courts have followed that ever since.

But that application asks for a throwaway line to carry far more weight than was intended. At most, PNB stands for the narrow proposition that a merging firm can’t justify harm in one market by pointing to unrelated benefits in another. It does not say a court must ignore everything outside a narrowly drawn market. Some enforcers have stretched it that far regardless, which is exactly what the states are leaning on.

The one question at the center of this deal, whether combining two old-line studios helps them compete with Netflix, Amazon, Apple, and YouTube, should still be given consideration. In a streaming market where content is king, the need to compete by constantly churning out new films and shows is what will incentivize the two studios to up their game and produce more quality features. The competitive dynamics cannot be ignored.

Neither can consumers and the simplest question is probably the most important – how, under this merger, does an ordinary viewer suffer harm? The states lead with statistics instead of a story about real harm, probably because the harm story is weaker than the math.

Now run it the other way. If the deal closes, you get one studio with the scale and incentive to invest and compete with the tech platforms. If it dies, you get two isolated companies struggling to keep up. Blocking this merger does not protect competition.

Antitrust law is supposed to protect competition and the people who benefit from it. A yardstick built for neighborhood banks cannot measure a business where a teenager’s phone competes with a movie theater.

There is also a bigger problem here. Federal enforcers ran a modern analysis and cleared this deal. Then a dozen state officials re-ran it under a 63-year-old rule in a courthouse they picked. If that can happen to any deal, federal clearance means very little.

Congress should make sure that deals of true national scope get one expert review under one modern standard, the way Europe has done for decades. Until then, Philadelphia National Bank will keep being used as a stand-in for the argument about consumer harm that the states cannot actually make.

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.