Some Feng Shui is indeed can be based on probability statistic.

For eg. House in front of T-junction has no good Feng Shui.
As probability of a car accidental hit into the house is high based on the mentioned location.
So in this issue, it makes a lot of sense.

While some Feng Shui can lead to 'feel good' effect which can create some placebo effect as well.

Placebo effect actually can be very powerful one which very difficult to explain in word and details.

Just like confidence issue, if one is confidence doing a thing, generally one can perform the task better than without confidence even though one is having enough skill already. Just like in sport, when a player is more confidence, he can delivery 110% of the ability, while confidence low time, schoolboy error also can happen on professional player.

So placebo effect could be identical to the confidence issue mentioned.

Yup.
No 44 is myth and hold not much ground. It is more about people 'feel good' factor or placebo effect. Feng Shui can have placebo effect on people as well.

But house at or infront of T-junction definitely has high probability being run into by a car. For eg. a car goes straight that break malfunction, then the probability the car will run into the house in front of T-junction is higher than the road side one.
But there is no guarantee the roadside house won't be run into by car, just the house infront of T-junction, or when car cornering time, the chance of accident is higher than ordinary straight road.
There is no guarantee the house in front of T-junction will run into by car as well. Just may be the probability is higher than the rest.

Yes, I shared you view on this, but undeniable, in term of statistically, the rate of accident occur with T-house or house at corner is definitely higher than ordinary one, it could be fractionanl or negligible, but surely it is higher.

At cornering time, the rate is accident is definitely higher than driving straight, due to mis-control, poor judgement etc issue.

Ok, I only focus on the probability issue, the rest issue could be more about placebo effect I posted earlier.

Don't look down on placebo effect, it could influence individual quite significantly eventually affect a lot of issue afterwards.