Page 1 of 1

Explain Perfect/Partial Perfect-N Schedule Like I'm Five

Posted: Mon Jan 05, 2015 9:14 am
by Vitamin K
So this one goes out to Stan Pope, though others are certainly welcome to chime in.

I'd like to give our emcee some copy to read to explain how we're doing scoring for our Derby, ahead of the races, so parents know what's going on. Given the short attention spans of most people, how can I most succinctly summarize PPN scheduling to them, in a way that lets them know what's going on, and why I'm doing a PN-finals after PPN-qualifiying round.

Re: Explain Perfect/Partial Perfect-N Schedule Like I'm Five

Posted: Mon Jan 05, 2015 9:32 am
by Stan Pope
Recite the criteria for PN, then recite the loosening of those criteria for PPN, especially about relaxation of total opposition balance and the inaccuracy that it introduces into point scoring.

PN Criteria:
1.Each car races in each lane the same number of times.
2.Each car races each opponent the same number of times.

PPN Criteria:
1. Each car races the same number of times in each lane (which implies that the number of races is a multiple of the number of cars)
2. Equality of opposition is optimized, i.e., no head-to-head matchup count exceeds another by more than 1.

The "loosening" is in #2. Instead of exactly balanced opposition, the opposition is almost equally balanced.

For example, a three lane, 8 car, 1 round chart each car will face all opponents except one once, but does not face one opponent.

Re: Explain Perfect/Partial Perfect-N Schedule Like I'm Five

Posted: Mon Jan 05, 2015 9:52 am
by Vitamin K
So maybe something like:

"For the qualifying rounds, we're using a computer algorithm that will give a schedule that ensures that each car races the same number of times in each lane, and attempts to match up opponents as equally as possible.

For the final round, we'll take a smaller set of the fastest cars and use a more restrictive algorithm that, in addition to the previous criteria, ensures that each car will race each opponent an equal amount of times."

Re: Explain Perfect/Partial Perfect-N Schedule Like I'm Five

Posted: Mon Jan 05, 2015 12:06 pm
by Stan Pope
Yes. You can add that there are more finalists than trophies because the imbalance may cause truly fastest to place lower because of match-ups in the preliminary competition.

My simulations have shown that the (number of trophies) fastest will almost always (99+% of the time) finish in at least (twice the number of trophies) place in a PPN chart.

Re: Explain Perfect/Partial Perfect-N Schedule Like I'm Five

Posted: Mon Jan 05, 2015 2:02 pm
by gpraceman
I summarize it as this:

1) Every racer will run in each lane of the track.
2) Every racer will run the same number of times.
3) All racers will face a variety of opponents.

Re: Explain Perfect/Partial Perfect-N Schedule Like I'm Five

Posted: Mon Jan 05, 2015 11:23 pm
by Darin McGrew
gpraceman wrote:I summarize it as this:

1) Every racer will run in each lane of the track.
2) Every racer will run the same number of times.
3) All racers will face a variety of opponents.
That's pretty much how I describe our system, which is basically a PPN system. Except we usually race each car twice in each lane.

And I usually phrase #3 something like "Each car will race against as many different opponents as possible." Technically, that's only true if there are 25+ cars (4 lanes = 3 opponents per race, x8 races = 24 opponents). But people seem to understand what I mean anyway.

Re: Explain Perfect/Partial Perfect-N Schedule Like I'm Five

Posted: Tue Jan 06, 2015 11:31 am
by Stan Pope
I think that these latter descriptions are okay for the five year old's. For some older who have heard about various final standings methods, go to the criteria I listed. Those criteria distinguish PN/PPN from such schemes as Stearns and Simple Lane Rotation, and they justify the assertions of fairness and accuracy.

Maybe you just mention that there is a more precise statement of criteria available for those who want to understand the why the PN/PPN method is used.