But still these repeats aren't that close to each other like for example dozen repeats are

A dozen is 12 numbers. The odds for a repeat of a single number starts with 1 number, then 2,3,4,5, etc. So the odds becoming increasingly high, with proportion to the number of unique numbers. So the odds are: 1/37, 2/37, 3/37 etc. So I don't understand your point of the above comment.

so it's very interesting to see this result

Exactly how is it interesting? If you include a previous number in your group, then of course the group is more likely to repeat that number.

It doesn't really matter how it happens, what matters that it happens

Yes it happens, but its no different to saying reds and blacks happen. It doesnt help knowing this.

I was just curious if you considered the possibilities of connections and dependencies after a decade of repeating that spins are independent.

Believe me I have an open mind to all possibilities. But I don't believe something without information to back it up. If I don't have information, I'll look for it. In this case, you provided information but what you thought was a connection is an oversight. I've been doing this a long time and have probably had more oversights than anyone here. There's nothing wrong with that. It's part of progress. But it's important to learn and progress when you understand the oversight.

even if we created it with cycles

You can't "create" something with roulette. You can't influence the spins. There's only whatever there is.

Beating roulette is about predicting spins with sufficient accuracy to overcome the unfair payouts. The question is HOW to do it. To me the most logical way is correlate the variables to the winning numbers. But I'm sure there are other ways not yet thought of.

after a decade of repeating that spins are independent

Until i see something that indicates otherwise, instead of everything that indicates there's no dependence, I'll naturally continue to have my reasonable belief.

you never no what will happen in the next spin, but in the next 10, 100 or 1000 spins you can predict certain events to happen with a high accuracy.

Maybe true, but I haven't yet found a solid way to demonstrate or prove this. There's lots we don't understand about the universe. But it makes sense to look at the road less traveled.

There are many examples of dependence but so what. It's still only statistics.

Its not at all dependence. The variables are different for each spin. Sometimes the variables are within ranges, and are predictable enough, for correlation. So you can predict spins with sufficient accuracy. For example, with dealer signature.

But most of you arent looking at tangible real-world variables. The way most of you are looking at it is like saying the weather in Australia is dependent on the weather in Moscow. In such a case, there is correlation, but not that you can realistically use one to predict the other. It's even worse with RNG.