Well, Suro is wrong in the specifics, so I'm hoping that he's ready to back down a little bit. In the main, he's correct -- the tariff calculations are ridiculous for a number of reasons in theory and at least one specific way empirically (i.e. the data does not exist). But that formula is not actually Trade Balance/Exports. It's more complicated. The formula also contained two greek letters, call them e and p, and those represent elasticity and pass-through of implemented tariffs. When you include those, it's actually a plausible model for the concept.
1. If you're trying to compute "what tariff is necessary to zero out the trade balance," and you are going to completely ignore exports, then you could get the job done without too much complexity by multiplying a) volume of imports, b) the amount that prices will go up in response to the tariff, c) the amount that imports will decline in response to the tariffs. And that's exactly the denominator in that silly equation, although they messed up in a couple of ways. The elasticity is itself a function of price level (and not just tariff), and pass through has to be a function of market concentration, but anyway.
Now, why did they divide that into the trade balance? I have no idea. I suspect it was a fudge to get the numbers higher. Trade balance/exports should always be less than 1 (I think) under realistic conditions, so it must have been used to get the numbers down. Like, the model probably spit out numbers in excess of 100%, and even Trump realizes that strains credulity.
2. The bigger problem is that it's conceptual nonsense. First, the US doesn't know what e or p is (in the equation above) so it's just making shit up. Second, there's no attempt to model the effects of doing all the tariffs at once (because it will produce way different results if you're tariffing everyone or only one). Third, zeroing the trade balance is not a desirable goal and doesn't even make sense. Fourth, there's no attempt to inquire what would happen to exports (which can't be held constant because they are specifically dependent on imports). So it's a load of bullshit.
3. But just FYI to avoid misinformation, I think Suro is wrong in his description of that formula. He's right that the numbers make no sense at all. The conservatives are going to jump on him about the technicality. If you discuss this with someone, you can now say that you've had the formula explained and it doesn't make any sense even if it's not technically what Suro described.