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

==============================================================================================================
 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= 290  acc=51.7%  naive CI [ 46%,  58%]  year-clustered CI [ 48%,  56%]  p_raw=0.2986  p_bonf=1.0000
  EV/EBITDA       n= 290  acc=54.5%  naive CI [ 49%,  60%]  year-clustered CI [ 49%,  61%]  p_raw=0.0710*  p_bonf=0.5678
  P/B             n= 276  acc=54.0%  naive CI [ 48%,  60%]  year-clustered CI [ 48%,  60%]  p_raw=0.1031  p_bonf=0.8245
  FCF yield       n= 252  acc=57.5%  naive CI [ 51%,  64%]  year-clustered CI [ 52%,  64%]  p_raw=0.0098***  p_bonf=0.0783*
  Earnings yld    n= 290  acc=51.7%  naive CI [ 46%,  58%]  year-clustered CI [ 48%,  56%]  p_raw=0.2986  p_bonf=1.0000
  EV/NOPAT        n= 290  acc=53.4%  naive CI [ 48%,  59%]  year-clustered CI [ 49%,  58%]  p_raw=0.1323  p_bonf=1.0000
  OE-DCF MoS      n= 290  acc=49.3%  naive CI [ 43%,  55%]  year-clustered CI [ 44%,  55%]  p_raw=0.6154  p_bonf=1.0000
  Brina Gap       n= 290  acc=54.1%  naive CI [ 48%,  60%]  year-clustered CI [ 48%,  59%]  p_raw=0.0884*  p_bonf=0.7069

==============================================================================================================
 POWER ANALYSIS (minimum detectable accuracy above chance)
==============================================================================================================
  P/E             n= 290  acc=51.7%  MDE_at_80%_power=57.3%    above MDE
  EV/EBITDA       n= 290  acc=54.5%  MDE_at_80%_power=57.3%    above MDE
  P/B             n= 276  acc=54.0%  MDE_at_80%_power=57.5%    above MDE
  FCF yield       n= 252  acc=57.5%  MDE_at_80%_power=57.8%    above MDE
  Earnings yld    n= 290  acc=51.7%  MDE_at_80%_power=57.3%    above MDE
  EV/NOPAT        n= 290  acc=53.4%  MDE_at_80%_power=57.3%    above MDE
  OE-DCF MoS      n= 290  acc=49.3%  MDE_at_80%_power=57.3%    above MDE
  Brina Gap       n= 290  acc=54.1%  MDE_at_80%_power=57.3%    above MDE

==============================================================================================================
 SAME COMPARISON, split by long-side / short-side
==============================================================================================================
  P/E             total n=290  acc=51.7% [ 46%,  57%] p=0.2986      LONG (cheap) n=145 acc= 51%     SHORT (expensive) n=145 acc= 52%
  EV/EBITDA       total n=290  acc=54.5% [ 49%,  60%] p=0.0710 *     LONG (cheap) n=145 acc= 55%     SHORT (expensive) n=145 acc= 54%
  P/B             total n=276  acc=54.0% [ 48%,  60%] p=0.1031      LONG (cheap) n=138 acc= 57%     SHORT (expensive) n=138 acc= 51%
  FCF yield       total n=252  acc=57.5% [ 52%,  63%] p=0.0098 ***     LONG (cheap) n=126 acc= 58%     SHORT (expensive) n=126 acc= 57%
  Earnings yld    total n=290  acc=51.7% [ 46%,  57%] p=0.2986      LONG (cheap) n=145 acc= 51%     SHORT (expensive) n=145 acc= 52%
  EV/NOPAT        total n=290  acc=53.4% [ 48%,  59%] p=0.1323      LONG (cheap) n=145 acc= 54%     SHORT (expensive) n=145 acc= 52%
  OE-DCF MoS      total n=290  acc=49.3% [ 43%,  55%] p=0.6154      LONG (cheap) n=145 acc= 52%     SHORT (expensive) n=145 acc= 47%
  Brina Gap       total n=290  acc=54.1% [ 48%,  60%] p=0.0884 *     LONG (cheap) n=145 acc= 57%     SHORT (expensive) n=145 acc= 51%

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