Kind of sad. Actually, both answers are wrong. There are a lot of patterns. I think the reason people say there aren't patterns is because there are exceptions to most of them, and because it's a verbose language containing many many patterns (e.g., it's a mnemonic chore even to remember the patterns), and finally because even when there is a rule that helps you to identify a kanji's reading, a kanji has multiple readings, so it's only useful for some of its readings. Every child learns the patterns in school, actually! And even foreigners, like me, can begin to guess, often correctly, what unknown kanji mean or say just by looking at them, eventually. This is the reason one of the first things students are supposed to learn when studying kanji is the radicals. Mainly, what you look for is called the r&p (radical and pronunciation) group. I don't have a good estimate, but I'm sure it comprises more than 30% and less than 80% of Japanese kanji. At a glance, it seems many kanji are r&p, anyway. One of the radicals creates the phonetic sound, and the other is a picture or figure of some kind that describes it. Very few Japanese kanji dictionaries describe this, so it's beneficial to learn Chinese if you want to recognize the patterns; otherwise, you need to just try to make the connections independently.
Patterns that you probably wouldn't be able to recognize until adulthood come from history. If you get the more in-depth kanji study dictionaries, which are meant for college students (get one at a Book Off, or you'll end up paying 200 bucks per each), give you the history of each kanji. Knowing the period means knowing the sound, basically. You will find that there is a much stronger overlap between the sounds kanji make and the radicals they contain when they are from the same period in Chinese history. You will also find that the meanings of radicals are closer if you can refine them by historical period. This is because possession of China changed