The Reality of Double Elimination
In preparing the double
elimination brackets used in the RaceManager program, every attempt has
been made to produce the fairest possible results.
Specifically, the heat order has been adjusted to equalize the waiting
time between runs and minimize the differential waiting time between two
cars in the same run. These not only produce fairer racing, but they also
help keep the race meet flowing smoothly. Click
here for more information on Hot/Cold pairing.
Cars going to the losing bracket are placed in a different section to
minimize the chance of their racing the same car again. However, in some
cases this is impossible to avoid. For instance it always happens when
the eventual 1st and 2nd place cars race each other before the final round.
A-B phase spacing, division merging and runoff heats are all done with
fairness in mind.
Given that the drivers and cars are consistent, that is, that they always
perform to the peak of their capability, we can say the following: DOUBLE
ELIMINATION RACING PROVIDES ACCURATE RESULTS 100% OF THE TIME FOR THE 1st
AND 2nd PLACE CARS ONLY.
The accuracy of the results for 3rd through 8th place gets increasingly
worse depending on the number of cars racing and the luck of the draw.
For example, with 20 cars in a division, depending solely on the luck of
the draw, the 3rd fastest car will place 3rd - 78% of the time, 4th - 16%,
and 5th - 6%. The 4th fastest car will place 3rd - 8%, 4th - 51% and 5th
- 36%. The 7th fastest car will place 3rd about 3% of the time, the 15th
car fastest car can be 6th about 2% of the time.
These numbers are based on 100 complete race meetings with each meeting
having a computer generated random heat assignment. The number of ways
to assign 20 cars is about 608,225,000,000,000,000. That's larger than
the number of seconds that have elapsed since the Big Bang!
The only way to get 100% accuracy in the 3rd place is to have triple
elimination.
SuperTimer II for the Soap Box Derby 800-654-2088 
