2026-05-15: $POET predicted +3.22%, did -11.02%

# Yesterday's Pattern Recap (May 15, 2026) Five picks went to market on May 15. The portfolio returned an average of −4.98% across the set. Directional calls landed 2 out of 5 correct. The single best performer was FIG, which yielded 6.28% against a 0.48% forecast—a modest beat. The worst position was POET, which fell 11.02% despite a predicted gain of 3.22%. POET served as yesterday's hero pick and also its largest miss. The pattern suggested a 3.22% upside move with a bearish directional label, a contradiction worth noting. Instead of gaining, the stock dropped 11.02 percentage points from forecast to actual—an absolute error of 14.24 points. This reversal dominated the day's performance drag. The position had been flagged as high-conviction on the chart pattern analysis, yet the actual move went in the opposite direction with material magnitude. This outcome illustrates the core calibration challenge: even readable patterns can fail sharply, and confidence in reversal or bounce scenarios can mask distribution risk. The 40% directional hit rate and the 500-basis-point portfolio headwind suggest the signal quality this week is below the threshold needed to outperform noise. The POET case specifically underscores that when predictions are directionally inverted by this margin, relying on pattern taxonomy alone leaves the methodology exposed. No adjustment to forward confidence levels is claimed here—only that yesterday's execution requires honest accounting before next week's picks are weighted.