That's an example of making your own luck. The team with the ball knows that need a basket in the final seconds, so they take the final gamble.
And what's to say that your favorite team didn't get to their finals spot with an equally large (though more subtle) amount of luck against the team they beat?
I recently read a book called The Drunkard's Walk: How Randomness Rules Our Lives. The basic premise of the book is that random factors have a huge influence in how life works out, especially in situations where you have incomplete knowledge.
Even if Pokemon had no critical hits, all moves always hit (barring specific moves like Fly that causes them to miss), the damage formula was 100% consistent, full paralysis didn't exist, and all the other elements of luck programmed into it were removed, luck would still play a significant part in any Pokemon match. Allow me to explain:
Let us assume a battle between two reasonable competent people. When you begin a Pokemon match, the only information you know about your opponent's team is the lead Pokemon, and in rare cases it's Ability. (The foe's Aerodactyl is exerting its Pressure!) The rest of your opponent's team, as well as their Items, Abilities, Movesets, Natures, EV spreads, and even their overall playstyle are entirely unknown to you. The exact same thing applies to your opponent. Until you know what your opponent is running, the pregame is largely guesswork on both player's parts. "Prediction" at this stage doesn't exist, it's largely just picking attacks at random and seeing what happens. Because both players don't know how the other's team is set up, there's a high probability that at this stage one of the players might inadvetantly make a wrong switch that devastates one of their team members. Blissey don't normally carry Grass Knot, so you might feel safe switching a Tyranitar in to beat her.
In the early game, with incomeplete information, the luck factor is huge, and cannot be avoided even if you could gain comeplete knowledge of your opponent's team beforehand. The opponent's playstyle would still be an unknown factor.
Even in the late game "prediction" is still largely hit or miss. Again, assuming that both players are reasonably close in skill, each player is roughly equally capable of outpredicting (and subsequently underpredicting) the opponent. The rate that one opponent would outguess that other in this kind of situation is largely a coin flip (or a dice roll, if you prefer). Gambles take place even at this level. You only need to read warstory comments saying "I left Gyarados in hoping he'd expect a switch and wouldn't use Thunderbolt..." or other things like that to realize that that decision was less of a prediction and more a calculated risk, i.e., a gamble, and therefore has an element of luck.
So, against anybody more skilled than a total newb, "prediction" is not fortelling the future, but more trying as best you can to either hedge your bets against the unexpected or making ballsy risks in the hope of a high reward.
In fact, in light of this new reasoning that I have just thought of, I will cease using the word "prediction". It's just not a proper description, and like "overcentralization" it's largely become an empty word. "Calculated risk", "educated guess", or evenly just simply "hunch" is a much more accurate word to use.
The point has been brought up before somewhere else, and I'll repeat it here. Skill in Pokemon does not come from eliminating luck, but in properly managing luck so that the odds are as in your favor as they can be. Luck will always be present.
So be happy, use your Paraflinching Jirachis/Togekisses and Super Luck Scope Lens Focus Energy Absols and let the luck flow. At the very least learn to deal with it, and stop saying that a match with one full paralysis and a critical hit has too much hax. I know I will.
