GPRM scheduling suggestion?

Debates and discussions on the various race scheduling methods that can be used and their fairness and accuracy in determining the winners.
Post Reply
User avatar
pwrd by tungsten
Master Pine Head
Master Pine Head
Posts: 723
Joined: Wed May 24, 2006 1:51 am
Location: Seattle, WA

GPRM scheduling suggestion?

Post by pwrd by tungsten » Mon Jan 26, 2015 4:04 am

Using GPRM. We have 19 cars and we have one very slow lane (slight exaggeration but it is noticeable to an untrained observer).

We want each car to race at lease 6 times and we want every car to run in each lane an even amount of times. In the end we will pick four cars for a final race off to determine top 3. I guess ideally every car would race every other car in the main race.

Any suggestions for race schedule for the main race and finals?

We need to assume that the timer may fail and we may have to score by position and not time.


W Racing!!!!

User avatar
Vitamin K
Master Pine Head
Master Pine Head
Posts: 980
Joined: Sat Apr 20, 2013 7:26 pm
Location: Montgomery County, MD
Contact:

Re: GPRM scheduling suggestion?

Post by Vitamin K » Mon Jan 26, 2015 5:18 am

pwrd by tungsten wrote:Using GPRM. We have 19 cars and we have one very slow lane (slight exaggeration but it is noticeable to an untrained observer).

We want each car to race at lease 6 times and we want every car to run in each lane an even amount of times. In the end we will pick four cars for a final race off to determine top 3. I guess ideally every car would race every other car in the main race.

Any suggestions for race schedule for the main race and finals?

We need to assume that the timer may fail and we may have to score by position and not time.
You could run a two-round partial-perfect-N race for the the initial rounds. If you use all four tracks, that would equate to eight runs per car, and if you used three (excluding the bad lane), you'd have six runs per car.

GPRM should be able to switch between point and time scoring on the fly, if I'm not mistaken.



User avatar
gpraceman
Site Admin
Site Admin
Posts: 4607
Joined: Fri Jun 20, 2003 12:46 am
Location: Highlands Ranch, CO
Contact:

Re: GPRM scheduling suggestion?

Post by gpraceman » Mon Jan 26, 2015 9:56 am

I would recommend the Perfect-N Type scheduling for both rounds.

If you want to get to the top 3 in the finals round, it would be better to advance at least twice that number of racers.

You can switch on the fly between times scoring and points scoring, if you encounter a timer issue during the race. The software tabulates points, even when in times scoring mode.


Randy Lisano
Romans 5:8

Awana Grand Prix and Pinewood Derby racing - Where a child, an adult and a small block of wood combine for a lot of fun and memories.

User avatar
pwrd by tungsten
Master Pine Head
Master Pine Head
Posts: 723
Joined: Wed May 24, 2006 1:51 am
Location: Seattle, WA

Re: GPRM scheduling suggestion?

Post by pwrd by tungsten » Tue Jan 27, 2015 4:46 pm

Thanks Everyone! Say we need a runoff for a tie between two cars? Race to two losses or a better approach?


W Racing!!!!

User avatar
gpraceman
Site Admin
Site Admin
Posts: 4607
Joined: Fri Jun 20, 2003 12:46 am
Location: Highlands Ranch, CO
Contact:

Re: GPRM scheduling suggestion?

Post by gpraceman » Tue Jan 27, 2015 4:51 pm

For a face off, I prefer a two run tiebreaker. The 2nd run, they swap lanes.

To do that in GPRM, you can use the "Test/Tie Break" feature and note down the times (or finish order) for the heats or you can create a 2nd data file that you enter the two racers into and then schedule them on just two lanes.


Randy Lisano
Romans 5:8

Awana Grand Prix and Pinewood Derby racing - Where a child, an adult and a small block of wood combine for a lot of fun and memories.

User avatar
Stan Pope
Pine Head Legend
Pine Head Legend
Posts: 6888
Joined: Sat Jul 05, 2003 7:01 pm
Location: Morton, Illinois
Contact:

Re: GPRM scheduling suggestion?

Post by Stan Pope » Tue Jan 27, 2015 9:27 pm

pwrd by tungsten wrote:Thanks Everyone! Say we need a runoff for a tie between two cars? Race to two losses or a better approach?
Yes, racing on alternating lanes until one of the two racers wins 2 heats on a row is a fair tie break. If the racers are more closely matched than the lanes, it may take way too long, though. Since the result may be determined by the quality of staging, the owner/driver of each car should stage his own car with close supervision to assure no accidental contact with the opponent's car!

You can also decide by 2 runs on alternating lanes and have the judges mark (on paper) the finish line separation ... how far behind was 2nd place. Then compare the marks on the paper. (This is analogous to comparing winning time margins, but without a timer!)


Stan
"If it's not for the boys, it's for the birds!"

User avatar
pwrd by tungsten
Master Pine Head
Master Pine Head
Posts: 723
Joined: Wed May 24, 2006 1:51 am
Location: Seattle, WA

Re: GPRM scheduling suggestion?

Post by pwrd by tungsten » Wed Jan 28, 2015 4:18 am

Thanks everyone!

thinking about the following:

Three lane track
1 race with 19 cars and PPN (2 rounds). Should I have two filler cars?

For the finals pick the top six cars (awarding 3 places). Have a drag race:

6th place -vs- 5th place.
winner races 4th place.
Winner races 3rd place.
winner races 2nd place.
Winner races 1st place.

Each race is both cars down the best two lanes on the track. then cars switch places and race again

If no winner repeat with each car getting a turn in each lane until there is a winner or break tie based on elapsed time if this goes too many heats....

Yay or nay?? :-)


W Racing!!!!

User avatar
Stan Pope
Pine Head Legend
Pine Head Legend
Posts: 6888
Joined: Sat Jul 05, 2003 7:01 pm
Location: Morton, Illinois
Contact:

Re: GPRM scheduling suggestion?

Post by Stan Pope » Wed Jan 28, 2015 1:15 pm

pwrd by tungsten wrote:Thanks everyone!

thinking about the following:

Three lane track
1 race with 19 cars and PPN (2 rounds). Should I have two filler cars?

For the finals pick the top six cars (awarding 3 places). Have a drag race:

6th place -vs- 5th place.
winner races 4th place.
Winner races 3rd place.
winner races 2nd place.
Winner races 1st place.

Each race is both cars down the best two lanes on the track. then cars switch places and race again

If no winner repeat with each car getting a turn in each lane until there is a winner or break tie based on elapsed time if this goes too many heats....

Yay or nay?? :-)
19 car ppn is not accurate enough to produce a quality ranking you plan for the drag race. Two fillers and run 21 car ppn (pn, actually <<< WRONG) improves accuracy and is maybe good enough to do the ranking. <<< WRONG

However, 6 runs on 3 lanes gives possible point spread of 6 through 18, i.e. 13 possible final scores. Since you have 19 racers, there will be some ties, usually around the central half of the scores. So you may not have a clear ranking of 1st thru 6th. I would have more confidence in a 7 car ppn (pn, actually) final because I have simulated this extensively to produce probabilities!
Last edited by Stan Pope on Wed Jan 28, 2015 9:19 pm, edited 1 time in total.
Reason: edited in "<<<WRONG" where I said something incorrect. See later post for correct info.


Stan
"If it's not for the boys, it's for the birds!"

User avatar
pwrd by tungsten
Master Pine Head
Master Pine Head
Posts: 723
Joined: Wed May 24, 2006 1:51 am
Location: Seattle, WA

Re: GPRM scheduling suggestion?

Post by pwrd by tungsten » Wed Jan 28, 2015 1:58 pm

Stan Pope wrote: 19 car ppn is not accurate enough to produce a quality ranking you plan for the drag race. Two fillers and run 21 car ppn (pn, actually) improves accuracy and is maybe good enough to do the ranking.

However, 6 runs on 3 lanes gives possible point spread of 6 through 18, i.e. 13 possible final scores. Since you have 19 racers, there will be some ties, usually around the central half of the scores. So you may not have a clear ranking of 1st thru 6th. I would have more confidence in a 7 car ppn (pn, actually) final because I have simulated this extensively to produce probabilities!
Does adding two filler cars make it pn?

I want to go to two lanes for the final because I am embarrassed about the slow lane and it could impact the outcome of the race. The rest of the race committee wants three lanes for the main race but will agree to two lanes for the finals...

Having inspected the cars already there are two cars that should pull away from the pack...


W Racing!!!!

User avatar
Stan Pope
Pine Head Legend
Pine Head Legend
Posts: 6888
Joined: Sat Jul 05, 2003 7:01 pm
Location: Morton, Illinois
Contact:

Re: GPRM scheduling suggestion?

Post by Stan Pope » Wed Jan 28, 2015 9:06 pm

pwrd by tungsten wrote:
Stan Pope wrote: 19 car ppn is not accurate enough to produce a quality ranking you plan for the drag race. Two fillers and run 21 car ppn (pn, actually) improves accuracy and is maybe good enough to do the ranking.

However, 6 runs on 3 lanes gives possible point spread of 6 through 18, i.e. 13 possible final scores. Since you have 19 racers, there will be some ties, usually around the central half of the scores. So you may not have a clear ranking of 1st thru 6th. I would have more confidence in a 7 car ppn (pn, actually) final because I have simulated this extensively to produce probabilities!
Does adding two filler cars make it pn?

I want to go to two lanes for the final because I am embarrassed about the slow lane and it could impact the outcome of the race. The rest of the race committee wants three lanes for the main race but will agree to two lanes for the finals...

Having inspected the cars already there are two cars that should pull away from the pack...
Oops! No! Just pulled up ppngen.htm and it says 21 is PPN. 6 runs 3 lanes 13 cars is PN (opponent balance), with 19 to 21 cars it is a PPN (opponent balance as good as possible but not perfect), so opponent balance is not as good. 19 cars 6 runs 3 lanes gives high probability that the 3 fastest are in the top 6 or 7 scores. I'd take 7 into either 3 run PN or 6 run CPN chart. 7 cars 3 lanes 6 runs results in all lane matchups being repeated with the cars exchanging lanes! Virtually all poor combinations of car speed and lane assignments disappear that way.

Image


Stan
"If it's not for the boys, it's for the birds!"

User avatar
pwrd by tungsten
Master Pine Head
Master Pine Head
Posts: 723
Joined: Wed May 24, 2006 1:51 am
Location: Seattle, WA

Re: GPRM scheduling suggestion?

Post by pwrd by tungsten » Thu Jan 29, 2015 1:19 am

Stan Pope wrote:
pwrd by tungsten wrote:
Does adding two filler cars make it pn?

I want to go to two lanes for the final because I am embarrassed about the slow lane and it could impact the outcome of the race. The rest of the race committee wants three lanes for the main race but will agree to two lanes for the finals...

Having inspected the cars already there are two cars that should pull away from the pack...
Oops! No! Just pulled up ppngen.htm and it says 21 is PPN. 6 runs 3 lanes 13 cars is PN (opponent balance), with 19 to 21 cars it is a PPN (opponent balance as good as possible but not perfect), so opponent balance is not as good. 19 cars 6 runs 3 lanes gives high probability that the 3 fastest are in the top 6 or 7 scores. I'd take 7 into either 3 run PN or 6 run CPN chart. 7 cars 3 lanes 6 runs results in all lane matchups being repeated with the cars exchanging lanes! Virtually all poor combinations of car speed and lane assignments disappear that way.

Image
Thanks Stan!


W Racing!!!!

Post Reply