Judge Kenny Smith out shines at Asheville-AI Judges Evaluation

WHC Admin Published: October 15, 2026 | Updated: October 15, 2026 4 minutes read

      In analysis of the judging results from the October 8–10, 2026, horse show in North Carolina Champoionships at Asheville, Fletcher, North Carolina, found that Kenny Smith’s placements were the closest to the official final results, while Dale Watts had the largest average difference. Paul Robbins fell between the two.
The evaluation compared the placings recorded by each of the three judges — Paul Robbins, Kenny Smith and Dale Watts — with the official final placements across 78 classes in the supplied results. The goal was to measure how closely each judge’s individual rankings aligned with the final outcomes.
Kenny Smith leads the comparison
Smith recorded the smallest average difference between his placings and the official final results, with an average discrepancy of 0.41 placing per horse. Robbins followed at 0.48, while Watts recorded 0.54.
Average difference from official final placements
Lower numbers indicate closer alignment.
Kenny Smith
0.41

Paul Robbins
0.48

Dale Watts
0.54

Smith’s average difference was approximately 0.13 placing lower than Watts’. Although that gap is relatively small on a placement scale, the results consistently put Smith first in this comparison.
How often did each judge match the final placing?
Another way to evaluate the results is to count how often a judge’s placement exactly matched the horse’s official final placing.
Judge Exact matches Placements compared Exact-match rate

Kenny Smith 263 376 69.9%
Dale Watts 247 374 66.0%
Paul Robbins 243 374 64.9%

Smith also led on exact matches, with his placement agreeing exactly with the final result in about 70% of the comparisons. Watts ranked second by this measure, while Robbins ranked third.
These figures illustrate an important distinction: the judge with the lowest average discrepancy is not necessarily the judge with the second-highest exact-match rate. Watts had more exact matches than Robbins, but Robbins’ nonmatching placements were, on average, closer to the final result. The two measurements provide complementary views of judging alignment.
What the numbers mean
A placement difference measures the distance between a judge’s ranking and the official final ranking. For example, if a horse finished second with a particular judge and third in the final results, the difference is one placing. If the judge placed that horse fifth and the final result was second, the difference is three placings.
The analysis averaged these differences across the available comparable placings. A lower average means a judge’s rankings tended to fall closer to the final outcomes, while a higher average indicates greater overall separation.
The comparison does not mean that Smith was necessarily the best judge in every class, or that Robbins and Watts were wrong whenever their placings differed from the final result. It summarizes the overall pattern across the available data.
Who was closest to the majority opinion?
Based on the official final placements used as the benchmark, Kenny Smith was the closest overall, Paul Robbins was next by average difference, and Dale Watts had the greatest average difference.
There is, however, an important limitation. The supplied results show the judges’ placings and the official final results, but they do not explicitly identify a majority vote for every horse. Therefore, the final placing is a practical benchmark for this analysis, not definitive proof of which judge agreed most often with the majority of the other judges.
Conclusion
The numerical comparison points to Kenny Smith as the judge whose placements most closely aligned with the official final results at the Fletcher show. He had both the lowest average placement difference and the highest exact-match rate. Paul Robbins had the second-lowest average difference, while Dale Watts had the highest average difference, despite recording more exact matches than Robbins.
Overall, the data favor Smith on both measures. The findings offer a useful statistical comparison of the three judges’ placements, while recognizing that alignment with final results is not, by itself, a complete assessment of judging quality or proof of majority agreement.