You should ask yourself how likely it is that anyone with the skills to effectively predict would post it publicly over time or share specifics about what they do. Jim Simons worked for years before he finally cracked it and yes - they predict. And no - they don't post on ET.
Most people who try to predict do it without any hard data, i.e., the market looks like a buy or a sell here. Quite a few people here actually seems to enter just based on a hunch and not back-tested signals/data.
Prediction requires data and a means to effectively calculate probabilities. Sure - at times the market may be a cointoss, but there certainly are times where certain patterns/outcomes are more likely than others.
Just a simple gap up open have a 73 % chance of the day closing positively net on the day. That's far more than 50 %. And yes, it's a rough prediction, but it's also just one parameter. If I add in one more parameter regarding the actual open - the probability of a net positive close on the day is 81 %.
Adding in more parameters - how much more predictable can the market possibly become?
Anything can happen and the improbable does happen from time to time. 2 weeks ago there was a pattern that's happened less than 10 times over the last 2000 days. And last week we had an Outside Day which is also among the less common patterns. Actually, we had two.
But most days are the regular up or down with range expansion with some small variations from day to day.