Exercise 2: The Sprint's Own ADR Threshold — Possible Solution
====================================================================

METHOD
------------------------------
Using the chapter's own formula (identical to Chapter 8's own
derivation): ADR_COST + wrong_prob*risk_reduction*R = wrong_prob*R,
solved for R.

RESULT
------------------------------
    Crossover reversal-cost threshold: 13.33 hours

CHECKING AGAINST THIS SPRINT'S OWN TWO REAL DECISIONS
------------------------------
    reversal_cost=40 (schema decision):    with_adr=8.00,  without_adr=12.00 -> ADR better
    reversal_cost=20:                       with_adr=5.00,  without_adr=6.00  -> ADR better
    reversal_cost=13.33 (the threshold):    with_adr=4.00,  without_adr=4.00  -> break-even
    reversal_cost=5:                        with_adr=2.75,  without_adr=1.50  -> no ADR better
    reversal_cost=1 (HTTP status decision): with_adr=2.15,  without_adr=0.30  -> no ADR better

Both of this sprint's own real decisions land cleanly on the correct
side of the 13.33-hour threshold: the 40-hour schema decision is well
above it (ADR clearly worthwhile), and the 1-hour HTTP status decision
is far below it (ADR clearly not worthwhile).

WHY THIS WORKS AS AN ANSWER
------------------------------
This confirms the threshold from Chapter 8's own Exercise 1 isn't just
an abstract number - it correctly sorts this specific sprint's own two
real decisions into the right buckets without needing separate
judgment calls for each. A team could apply this exact threshold
mechanically during sprint planning ("would reversing this cost more
than about 13 hours of work?") to decide, in seconds, whether a given
decision deserves an ADR - turning Chapter 8's own abstract finding
into a genuinely usable sprint-planning heuristic.