Records (VALID rows with 5y forward returns): 451

================================================================
 MATCHED-PAIR ANALYSIS
 Within fiscal year, control for fundamentals, isolate price
================================================================

  L1 LOOSE  — ROIC ±10pp, g_f ±10pp, any sector
    Pairs tested: 2521
    Gap-differential predicts: hits=1284/2521 = 50.9% CI [49%, 53%] p=0.1798 
    MoS-differential predicts: hits=1298/2521 = 51.5% CI [50%, 53%] p=0.0703 *

  L2 MEDIUM — ROIC ±5pp, g_f ±5pp, any sector
    Pairs tested: 908
    Gap-differential predicts: hits=447/908 = 49.2% CI [46%, 53%] p=0.6907 
    MoS-differential predicts: hits=451/908 = 49.7% CI [46%, 53%] p=0.5918 

  L3 STRICT — ROIC ±3pp, g_f ±3pp, SAME sector
    Pairs tested: 59
    Gap-differential predicts: hits=29/59 = 49.2% CI [37%, 61%] p=0.6026 
    MoS-differential predicts: hits=30/59 = 50.8% CI [37%, 63%] p=0.5000 

================================================================
 + GAP-DIFFERENTIAL FLOOR (require |Gap_A − Gap_B| > 5pp)
 Tests whether the framework predicts when the price signal
 is loud, not just whenever fundamentals are matched.
================================================================

  L1 LOOSE  + |ΔGap| > 5pp
    Pairs tested: 1432
    Gap-differential predicts: hits=736/1432 = 51.4% CI [49%, 54%] p=0.1514 
    MoS-differential predicts: hits=712/1432 = 49.7% CI [47%, 52%] p=0.5940 

  L2 MEDIUM + |ΔGap| > 5pp
    Pairs tested: 460
    Gap-differential predicts: hits=219/460 = 47.6% CI [43%, 52%] p=0.8582 
    MoS-differential predicts: hits=215/460 = 46.7% CI [42%, 51%] p=0.9259 

  L3 STRICT + |ΔGap| > 5pp
    Pairs tested: 21
    Gap-differential predicts: hits=7/21 = 33.3% CI [14%, 52%] p=0.9608 
    MoS-differential predicts: hits=10/21 = 47.6% CI [29%, 67%] p=0.6682 

  L1 LOOSE  + |ΔGap| > 10pp
    Pairs tested: 641
    Gap-differential predicts: hits=337/641 = 52.6% CI [49%, 56%] p=0.1031 
    MoS-differential predicts: hits=318/641 = 49.6% CI [46%, 54%] p=0.5937 

================================================================
 INTERPRETATION
================================================================
 - If Gap-differential accuracy > 50% with significance after
   matching, the Brina Gap adds price-discovery value beyond
   what fundamentals already predict.
 - If accuracy ≈ 50%, the Gap signal is noise once fundamentals
   are controlled for; the framework's univariate accuracy
   reflects fundamental differences, not the price signal.
 - Compare Gap accuracy vs MoS accuracy at each level — they
   should diverge if our earlier finding (MoS is at chance) is
   robust within matched pairs.
