Records (VALID rows with 5y forward returns): 327

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

  L1 LOOSE  — ROIC ±10pp, g_f ±10pp, any sector
    Pairs tested: 1242
    Gap-differential predicts: hits=634/1242 = 51.0% CI [48%, 54%] p=0.2391 
    MoS-differential predicts: hits=631/1242 = 50.8% CI [48%, 54%] p=0.2949 

  L2 MEDIUM — ROIC ±5pp, g_f ±5pp, any sector
    Pairs tested: 412
    Gap-differential predicts: hits=213/412 = 51.7% CI [47%, 57%] p=0.2610 
    MoS-differential predicts: hits=206/412 = 50.0% CI [45%, 55%] p=0.5196 

  L3 STRICT — ROIC ±3pp, g_f ±3pp, SAME sector
    Pairs tested: 21
    Gap-differential predicts: hits=9/21 = 42.9% CI [24%, 62%] p=0.8083 
    MoS-differential predicts: hits=12/21 = 57.1% CI [33%, 76%] p=0.3318 

================================================================
 + 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: 731
    Gap-differential predicts: hits=369/731 = 50.5% CI [47%, 54%] p=0.4122 
    MoS-differential predicts: hits=361/731 = 49.4% CI [46%, 53%] p=0.6442 

  L2 MEDIUM + |ΔGap| > 5pp
    Pairs tested: 221
    Gap-differential predicts: hits=108/221 = 48.9% CI [43%, 56%] p=0.6567 
    MoS-differential predicts: hits=108/221 = 48.9% CI [43%, 56%] p=0.6567 

  L3 STRICT + |ΔGap| > 5pp
    Pairs tested: 7
    Gap-differential predicts: hits=2/7 = 28.6% CI [0%, 57%] p=0.9375 
    MoS-differential predicts: hits=4/7 = 57.1% CI [14%, 86%] p=0.5000 

  L1 LOOSE  + |ΔGap| > 10pp
    Pairs tested: 350
    Gap-differential predicts: hits=183/350 = 52.3% CI [47%, 57%] p=0.2114 
    MoS-differential predicts: hits=178/350 = 50.9% CI [45%, 56%] p=0.3947 

================================================================
 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.
