Right, you don't know. Circumstantial evidence is not proof, unless it happens often enough, to rule out the possibility of it being a random occurrence.
Anyone who has ever been in the auto body repair business can tell you, that your "married" to every vehicle you repair.
The slightest noise, vibration, or running issue, and then vehicle is back at your shop, with the customer insisting that it was caused by the accident. They really want to believe this, because they feel it will be a covered by insurance, and no cost repair for them. So, the particular problem, amount of time, and circumstances don't ever matter, you still hear about it.
We also had a large percentage of repeat customers, whom we often knew rather well. They always wanted to talk about their cars. And we also made follow up calls, to check with customers to verify everything was okay.
You can't imagine how many unrelated, worn out parts, I have replaced over the years, often at our expense, to satisfy customers.
None of the above has ever involved the suspicion, or replacement, of a defective computer.
The reason we did not, as a rule, disconnect a customers battery, was NOT because we were lazy. We found they did not appreciate losing their ratio station presets, and clock setting. Years ago, many vehicles had radio's with strange clock setting procedures, that occasionally required a lot of wasted time, trying to figure them out, after disgruntled customers could not.
This decision to not disconnect, was made above my pay grade, and there never was any evidence that it had caused any problems.
I am not suggesting that people do not disconnect the battery, or that a welder cannot somehow damage a computer. I am trying, now in vane, to show that computers, and or electrical problems, as a result of not disconnecting the battery, before welding, was not a problem, in a very large sample, for us.