From a theoretical standpoint, it's pretty much always better to go first.
It's a lot easier to draw out on a piece of paper, but the basic explanation is this: If you go first, you get one turn with no attacks, followed by an opponent's turn with attacks, followed by your turn with attacks. At this point, you have a 100% advantage over your opponent--because you have had two turns (one with attacking, one without), and your opponent has only had one turn (with attacking). On the next turn, you will be at a slight disadvantage--you will have still had two turns (one with attacks, one without), and your opponent will have had two turns (both with attacking). That means every other turn you will be at a 100% advantage, and the turns in between you will be at a slight disadvantage. On the flip side, if you choose to go second, this system works in the opposite way--every other turn you are at a 100% disadvantage, and every turn in between you are at a slight advantage. So basically, you're weighing a 100% advantage + a slight disadvantage against a 100% disadvantage + a slight advantage. Therefore, based on this, going first is always better.
Of course, without even going into that kind of depth, it's still pretty clear that you should always choose to go first. After all, you have no idea what you are going to start with nor what your opening hand will be; you could choose to go second and not be able to attack anyway because of a bad start. The only situation where going second is the better option is where getting an attack before your opponent is going to allow you to set up faster than them--in short, when the attack is more important than getting things like the first attachment or establishing your field. The only time this will ever happen is when there are two aggressive decks facing each other, and without knowing what you are playing against beforehand, you can't really decide to go second based on this. Once again, you also can't guarantee you'll even get an attack if you decide to go second, meaning literally the only situation you should go second is if not getting the first attack will put you at a massive disadvantage--something that will rarely, if ever, be the case.
tl;dr, just go first, always. The situations where going second is the better option are practically nonexistant, and because you have to decide before you see your opening hand, you will never be informed enough to know that going second is really the better play.