Eyes is based solely on 'aahhhh scary, 1 in a million is still a chance!" That's not evidence or statistics, it's literally superstition and hearsay. How do we find out how dangerous it really is? And we do know this could be mitigated with perf plate or slotted oculars.
I just wanted to briefly circle back to this before it's forgotten, lost to the annals of time on the forum.
I was just reading an article that reminded me of this discussion.
The above is really not a valid argument from a scientific/engineering standpoint. Generally speaking, seemingly improbably events are often more probable than assumed, particularly with repetition.
For instance, people buy lottery tickets all the time. Your chances of winning are extremely low, and yet people win all the time, due to the rate at which they are purchased. Your chances of dying in a helicopter are very low on any given day, but helicopters can and do crash. Your chances of having it happen to you over the course of your lifetime increase dramatically, the longer you live. This is even more true with driving cars, where the accident rate is quite high.
Some of the most tragic events in human history were extremely unlikely. A great example is the Titanic, in which over a dozen things went wrong, any one of which you might say was only a 1 in 10,000 chance of occurring. Multiply those out, and the chances of that ship going down were astronomically small. And yet, that's exactly how significant tragedies occur-- It can take a veritable orgy of improbable events happening all at once, or in succession. Sometimes the reality behaves as though the probabilities add, rather than multiply.
The biggest mistake in statistics and probability is to assume that patterns won't emerge. They can and do, even from random noise.
In a case like this, the potential damage from an eye-slot thrust is enormous (fatality is highly possible since the bone behind the eye sockets is not very strong), whereas the means to prevent it is relatively easy. Perf-plate is one solution. Swords that are too thick is another. It just comes down to which is more cost-effective or available at the time. But saying that the probability is low would not make me want to fight with a combination of sword and visor that won't 100% stop a thrust. I might never get struck through that visor, but if we accept it as standard practice, then someone will be injured eventually.