Premier League Predictor: Astro — Chapter 6, Exercise 1
====================================================

TASK
Explain, with a concrete example, why scorePrediction() checks for an
exact score match before checking for a correct-result match, and
what would go wrong if the outcome check ran first and returned
immediately on a match.

SOLUTION

Every exact score match is automatically an outcome match too: if a
prediction of 2-1 exactly matches an actual result of 2-1, then the
outcome derived from that prediction (home win) also automatically
matches the outcome derived from the actual result (also home win).
There's no such thing as an exact score match that isn't also an
outcome match — the reverse is what can differ (3-0 predicted against
a 1-0 actual result matches on outcome only, not on score).

Because of that relationship, if outcome() were checked first and the
function returned immediately on a match, the exact-score branch would
never be reached for a genuinely exact prediction. Concretely: a user
predicts 2-1, and the match finishes 2-1. Under an outcome-first
version of scorePrediction, the very first check would compute
outcome(2, 1) === outcome(2, 1), find that HOME_WIN === HOME_WIN, and
return POINTS_CORRECT_RESULT (10) right there — the function would
never get as far as comparing the actual scorelines, so the 40-point
exact match would silently be scored as a 10-point outcome match
instead. Every single prediction that got the exact score right would
be under-scored by 30 points, every time, with no error or warning of
any kind.

This is why the real order in the chapter's own implementation checks
the exact score first and returns immediately on that branch: it's
the only way to distinguish "this earned the full 40 points" from
"this only earned the smaller 10-point consolation" before the more
general outcome check ever gets a chance to fire.

WHY THIS WORKS AS AN ANSWER
----------------------------
It states the precise logical relationship between an exact match and
an outcome match (the former always implies the latter, never the
reverse), works through a concrete scoreline pair to show the actual
under-scoring bug that a reordered check would produce, and quantifies
the real consequence (every exact prediction losing 30 points) rather
than only asserting that the order "matters."