Scoring Analysis from 2011 Finals
Lisbeth and I enjoy statistical analysis and looking at probability analysis. We both were intrigued with the question posed about possible changes in the way we select who moves on to the semi-finals at the National Finals. So, we sat down and did a quick look at this year’s data.
We went back and looked at the preliminary scores to discern how many handlers went forward on each of the preliminary days. Because of ties this year, 43 handler/dog teams moved forward into this year’s semi-finals (versus the prescribed 40). We found that the following number moved forward on Tuesday: 10; Wednesday: 11; Thursday: 9; and, Friday: 13.
We believe that this amount of discrepancy by day would not support the notion that one day was significantly more difficult or easier than another. To analyze this further, we totaled the number of handlers who received no scores (RTs/DQs)—Tuesday: 9; Wednesday: 7; Thursday: 9; Friday: 9. These figures also seem to support the notion that one day was not any harder or easier than another.
We also looked at which handlers would have qualified for the semi-finals if the ten highest scores per day where selected (as is currently being proposed). For this year, that would mean that on Tuesday qualifiers would remain the same. On Wednesday, one qualifier would be omitted (in this case Noel Williams/Lad). On Thursday, the ten highest scores would have included Vergil Holland/Scott with a 127. The ten highest scores on Friday would have knocked out Vickie Close/Hanna with a 130, Louanne Twa/Meg with a 131, and, possibly either Don Helsley/Tag with a 133, or, Rose Anderson/Shen with a 133 (depending on new rules would propose to handle ties). Remember that if you are choosing teams by the day they competed, tie scores would not be by overall competition, but by the day on which the tie occurred. So, each of the 130s that happened on the other days would only count if they were part of the top ten on THAT particular day. (Example: Noelle Williams/Lad; Mary Thompson/Beauty; Roy Taber/Craig; and, Mike Hanley/Moss received a 130 on the “right” days.) Another way to look at these data are that the team Holland/Scott placed 49th after the preliminary runs. They would compete in semi-finals, while teams that were ranked 34th, 35th, 36th, 38th, and 43rd would potentially be out.
Further examination reveals that D. Keeton/York (129), C.Riley/Nan(128), D. Boyce/Tink(129), T. Fleming/Bill(128), and, Emil Luedcke/Spot(127)–although scoring equal to or higher than Vergil Holland/Scott (127)–achieved their scores on the “wrong” day. We are not sure who would want to tell those people that they did not make the semi-finals, while Holland/Scott did. (See tables below.)
We are not convinced that if you were to ask someone who ran on Friday (which had the most qualifying runs this year), that they would say that Friday was a lucky draw and an easy day to run on compared to the other 3 days.
As a side note, Lisbeth and I have also looked at the data over several years from Meeker (another trial that covers several days of runs that move into the semi-finals). We have observed that day of the week has not generally been a major factor over who qualifies for the semi-finals.
In conclusion, we would suggest that before the Board of Directors changes the method for choosing the semi-finalists at National Finals, an analysis such as we have computed for 2011 be done on data from the 2007 through 2010 years. We have never gone to Nationals when they were held in the East. We think it is important to analyze data from those trial settings before considering a rule change.
I hope this makes sense.
Linda (and Lisbeth)
PS We have nothing against Vergil Holland. This is just how the numbers turned out! J
Table Example: A look at top ten from each of the preliminary runs on each of the days:
**Note: Green—indicates the handler would have been in if a 10/day had been used. Red—indicates that the handler team would have been out. Yellow—indicates that the handler team might have been out depending on how ties would be handled.
Tuesday (10 qualifiers)—RT (1), DQ(8) total 9
Wednesday (11 qualifiers)– RT(4), DQ(3) total 7
Thursday (9 qualifiers)– RT(4), DQ(5) total 9
Friday (13 qualifiers)– RT(5), DQ(4) total 9
|35 (Tie-Possible out)||D.Helsley/Tag||133||AM|
|36 (Tie-Possible out)||R.Anderson/Shen||133||PM|