I agree that pika's small shield reels some opponents in much closer than they should be (I wrote something about it a few pages back) but it's still easy to outspace against for smart players because of pika's crappy range.
Shieldbaiting is a scary thing to gamble with since pika in shield is a really bad position in general, but you can use it if your opponent has an easily exploited shield pressure game. In these cases you can get away with turnaround shield > uair oos all day. You could also shieldbait if you already know your opponent reacts to shield a certain way. This is only if they don't adapt to your shieldbaits well, because if they do you should just reserve it for when you're in defense since being stuck in shield isn't worth starting the mixup rps game imo (esp since the opponent has so many more options than you).
I feel that (especially for oos options) shieldgrabbing itself shouldn't ever be based off creating a setup to read an approach on your shield but rather out of reaction during defense to bad spacing/timing, missed lcancels, etc. Sure, shield sdi will improve your chances of closing in on them but trying to read whether they'll do a crossup/close/far spaced shffl on your shield isn't as effective as simply reacting to their spacing. Also it's stupid easy to punish against, only reinforced by the fact that so many players (even casual players) know how to space themselves just so that they can't be shieldgrabbed since it's one of the first oos defensive options people learn.
As for reading in general, IMO, understanding why your opponent does things the way he does is as important as memorizing habits, if not more important. For example (going off what Spyro said about teching) habit-wise you can say 'he always techs in place, but when I punish him he starts teching in, and his memory of my option coverage spans about <#> choices' but interpretation could be like 'teching in place = more worried about regaining control = player is aggro, teching in = more worried about survival = survival di = easier to combo, teching out/away = more worried about escaping = escape di = easier to kill, no tech = really risky attempt to escape a read/accident = player is panicking mad hard.'
Of course, thought like this varies from person to person and the way I wrote it is really linear, but I think initial tech habits (and habits in general prior to adaptation) give a good idea on how the opponent prioritizes his gameplay. Understanding habits is more important than memorizing them when you're reacting instead of predicting, and can segue you from reaction into prediction more easily later in the match (where both 'why' and 'what' will help immensely).
...or you can just dd space outside danger zone and always win