Records (VALID rows with 5y forward returns + all primitives): 260

==============================================================================================================
 HEAD-TO-HEAD: 8 value signals on identical data
 Tercile method: top-1/3 cheap → predict outperform, bottom-1/3 expensive → underperform
 Bonferroni correction: p_bonf = min(p_raw × 8, 1.0)
==============================================================================================================
  P/E             n= 166  acc=47.0%  naive CI [ 39%,  55%]  year-clustered CI [ 41%,  52%]  p_raw=0.8033  p_bonf=1.0000
  EV/EBITDA       n= 166  acc=49.4%  naive CI [ 42%,  57%]  year-clustered CI [ 43%,  57%]  p_raw=0.5920  p_bonf=1.0000
  P/B             n= 154  acc=44.8%  naive CI [ 37%,  53%]  year-clustered CI [ 38%,  51%]  p_raw=0.9147  p_bonf=1.0000
  FCF yield       n= 140  acc=55.7%  naive CI [ 48%,  64%]  year-clustered CI [ 48%,  64%]  p_raw=0.1024  p_bonf=0.8189
  Earnings yld    n= 166  acc=47.0%  naive CI [ 39%,  55%]  year-clustered CI [ 41%,  52%]  p_raw=0.8033  p_bonf=1.0000
  EV/NOPAT        n= 166  acc=52.4%  naive CI [ 45%,  60%]  year-clustered CI [ 44%,  59%]  p_raw=0.2935  p_bonf=1.0000
  OE-DCF MoS      n= 166  acc=53.0%  naive CI [ 46%,  61%]  year-clustered CI [ 49%,  58%]  p_raw=0.2425  p_bonf=1.0000
  Brina Gap       n= 166  acc=60.2%  naive CI [ 52%,  67%]  year-clustered CI [ 50%,  69%]  p_raw=0.0051***  p_bonf=0.0409**

==============================================================================================================
 POWER ANALYSIS (minimum detectable accuracy above chance)
==============================================================================================================
  P/E             n= 166  acc=47.0%  MDE_at_80%_power=59.7%    above MDE
  EV/EBITDA       n= 166  acc=49.4%  MDE_at_80%_power=59.7%    above MDE
  P/B             n= 154  acc=44.8%  MDE_at_80%_power=60.0%    above MDE
  FCF yield       n= 140  acc=55.7%  MDE_at_80%_power=60.5%    above MDE
  Earnings yld    n= 166  acc=47.0%  MDE_at_80%_power=59.7%    above MDE
  EV/NOPAT        n= 166  acc=52.4%  MDE_at_80%_power=59.7%    above MDE
  OE-DCF MoS      n= 166  acc=53.0%  MDE_at_80%_power=59.7%    above MDE
  Brina Gap       n= 166  acc=60.2%  MDE_at_80%_power=59.7%  ✓ above MDE

==============================================================================================================
 SAME COMPARISON, split by long-side / short-side
==============================================================================================================
  P/E             total n=166  acc=47.0% [ 39%,  54%] p=0.8033      LONG (cheap) n= 83 acc= 47%     SHORT (expensive) n= 83 acc= 47%
  EV/EBITDA       total n=166  acc=49.4% [ 42%,  57%] p=0.5920      LONG (cheap) n= 83 acc= 49%     SHORT (expensive) n= 83 acc= 49%
  P/B             total n=154  acc=44.8% [ 37%,  53%] p=0.9147      LONG (cheap) n= 77 acc= 48%     SHORT (expensive) n= 77 acc= 42%
  FCF yield       total n=140  acc=55.7% [ 48%,  64%] p=0.1024      LONG (cheap) n= 70 acc= 56%     SHORT (expensive) n= 70 acc= 56%
  Earnings yld    total n=166  acc=47.0% [ 39%,  54%] p=0.8033      LONG (cheap) n= 83 acc= 47%     SHORT (expensive) n= 83 acc= 47%
  EV/NOPAT        total n=166  acc=52.4% [ 45%,  60%] p=0.2935      LONG (cheap) n= 83 acc= 52%     SHORT (expensive) n= 83 acc= 53%
  OE-DCF MoS      total n=166  acc=53.0% [ 45%,  61%] p=0.2425      LONG (cheap) n= 83 acc= 54%     SHORT (expensive) n= 83 acc= 52%
  Brina Gap       total n=166  acc=60.2% [ 52%,  67%] p=0.0051 ***     LONG (cheap) n= 83 acc= 61%     SHORT (expensive) n= 83 acc= 59%

==============================================================================================================
 INTERPRETATION GUIDE
==============================================================================================================
 - Tercile-based signal test is the conventional factor-portfolio framework.
 - All signals tested on the same set of underlying observations.
 - p_raw: naive binomial p-value (treats observations as i.i.d.).
 - p_bonf: Bonferroni-corrected for 8 simultaneous tests.
 - naive CI: per-observation bootstrap (anticonservative; assumes independence).
 - year-clustered CI: per-year bootstrap (proper inference under within-year
   correlation in stock returns; the methodologically correct interval).
 - MDE = minimum detectable effect size at α=0.05 (one-sided) and β=0.20.
   Signals with acc < MDE are underpowered at this sample size.
 - Practitioner takeaway: a signal that survives Bonferroni AND has a
   year-clustered CI excluding 50% is robust to standard reviewer objections.
