Mark Zarb’s 1967 AFL card set graded out at 3.9 resulting in a “Very Good” rating. This is a borderline “Excellent” set that scored several “Excellent” ratings. Once again, overall yardage is skewed; it seems to be a trend with his earlier created sets. A total of 26 categories were evaluated.
Mark Zarb’s 1965 AFL card set graded out at 3.5 resulting in a “Very Good” rating. This set produced “Excellent” results with regards to fumbles, touchdowns and field goals. Uncharacteristically, this set received a “Poor” rating in total net yardage. A total of 26 categories were evaluated.
Having been in the test and evaluation business for the past 25 years, I thought it would be fun to create a template to measure the “statistical accuracy” of a card set. I will use a five-tier (Excellent, Very Good, Good, Fair and Poor) rating system to individually evaluate team wins, offensive and defensive categories, if applicable. I will compare the replay’s league average against the actual league average per category, assign a weighted score (i.e., Excellent is worth 5 points, Very Good is 4 points, etc.), sum and average to determine the overall rating. For example, if the league average for first downs was 36.3 per game and the replay averaged 34.2 resulting in a difference (+/-) of 2.1. When compared against the below criteria, this category would be rated as “Fair” and assigned 2 points.
|First Downs||Excellent – 0.0 – 0.5|
|Very Good – 0.6 – 1.0|
|Good – 1.1 – 2.0|
|Fair – 2.1 – 3.0|
|Poor – 3.1 or more|
My criteria for each category is stringent, however, it’s in line with my “pursuit of the perfect replay”.