N-elim format lane selection

Debates and discussions on the various race scheduling methods that can be used and their fairness and accuracy in determining the winners.
Post Reply
dfscott
Journeyman
Journeyman
Posts: 40
Joined: Fri Dec 03, 2010 1:34 pm
Location: Atlanta, GA

N-elim format lane selection

Post by dfscott » Sat Dec 04, 2010 9:43 pm

My Den Leaders were very resistant to a non-elimination format race, so I'm looking at doing a quad-elimination format for each rank (will take longer, but it shouldn't be any longer than last year's single elim, best 2 out of 3 race).

However, looking at Stan's lane selection methods, I'm worried that C is going to frustrate a lot of younger cubs. They make it to the front of line and are told to go to the back. Then they get back up to the front and (might) have to go to the back again! I only have between 7 and 11 cubs in each group, but there's still a decent chance to be sent to the back multiple times.

So what's the difference in "fairness" between B and C? I think I could sell "you're in the next round" a lot easier, even if they did bump more than once.



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: N-elim format lane selection

Post by Stan Pope » Sat Dec 04, 2010 11:41 pm

dfscott wrote:My Den Leaders were very resistant to a non-elimination format race, so I'm looking at doing a quad-elimination format for each rank (will take longer, but it shouldn't be any longer than last year's single elim, best 2 out of 3 race).

However, looking at Stan's lane selection methods, I'm worried that C is going to frustrate a lot of younger cubs. They make it to the front of line and are told to go to the back. Then they get back up to the front and (might) have to go to the back again! I only have between 7 and 11 cubs in each group, but there's still a decent chance to be sent to the back multiple times.

So what's the difference in "fairness" between B and C? I think I could sell "you're in the next round" a lot easier, even if they did bump more than once.
There are strong reasons for "C" vs "B":
1. Less "gamesmanship" to be at the back of the line and sometimes get fewer opponents. (This is more important when racing 2 lanes than 3 lanes like we do.)
2. The Cub at the end of the line gets to the front twice as fast!
3. "C" improves the randomization of match-ups relative to the "self ordering" that happens when the line id formed.

"B" diminishes the value of "end of line gamesmanship", but does not negate it to my satisfaction. That was the main reason I built "C" after watching a couple years of district races using "B". :)

Running small groups like that, be sure to run some simulations to see what the racing will be like ... how many heats.

You could stop racing when all score groups reach 1, but I like to keep the suspense for 1st place to be left for the last heat, so I use the ladder where the survivors, starting with the worst scoring survivor get a chance to "knock off" the one above by winning two straight heats against him (Draw for lane each heat). Why twice instead of once? Well, if they only raced once, they would have the same number of losses! On our good district tracks, we only occasionally see the standings change during that final ladder, but, once in a while they do change and a car may climb from 5th up to 4th or 3rd or, maybe, even 2nd. I've never seen one climb from 5th (4 losses during the elimination stage) to 1st at the end of the finals ladder, but it is possible. And if the standings don't change during the final ladder, well, it confirms the standings.


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

dfscott
Journeyman
Journeyman
Posts: 40
Joined: Fri Dec 03, 2010 1:34 pm
Location: Atlanta, GA

Re: N-elim format lane selection

Post by dfscott » Sun Dec 05, 2010 12:47 am

Stan Pope wrote: 2. The Cub at the end of the line gets to the front twice as fast!
Ah, good point! I hadn't thought about it that way!
Stan Pope wrote: Running small groups like that, be sure to run some simulations to see what the racing will be like ... how many heats.

You could stop racing when all score groups reach 1, but I like to keep the suspense for 1st place to be left for the last heat, so I use the ladder where the survivors, starting with the worst scoring survivor get a chance to "knock off" the one above by winning two straight heats against him (Draw for lane each heat).
Yep, I was a bit worried about that as well. I ran the simulations (assuming full participation from everyone in the pack) and came up with this:

Webelos - 11 cars - 57 heats
Bears - 11 cars - 57 heats
Wolves - 7 cars - 35 heats
Tigers - 9 cars - 45 min
Pack Finals (top 3 from previous) - 12 cars - 60 heats

Definitely a full day of racing. I'm hoping I can run one 2-lane heat every 60 seconds -- not sure how optimistic that is. If so, we're looking at over 4 hours but hey -- it's a FUN 4 hours.

By the way, I am planning to use your KO method for the reasons you mentioned.

And thanks again for all your help!



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: N-elim format lane selection

Post by Stan Pope » Sun Dec 05, 2010 2:14 am

From the simulations, you can see how many Cubs leave without winning a race against their peers. The number is not very large. If you want everyone to leave with a contested win (i.e. excluding bye wins), there are two ways that I know of to do so: Each racer that is eliminated with no wins ...
1. gets to race the Cub Master's Super Sleek (and very slow) car, or
2. gets to race against the other winless, and are eliminated from this "single elim" by winning. Last boy standing races the Cub Master's SS(avs) car.

While #2 at first appeared attractive to me, it does single out the slowest of the slow. So, I think I prefer #1. But, that CM car had better be slowwwww!

If you have a pack Mom or Dad who is willing and capable, they could spend a few minutes with each boy to help fix problems with the car to speed it up before racing the CM car. Could be good a "learning experience" for the youngster. A pocket full of "worked axles" and a ProBody tool might be all that is needed to rescue the car.
Last edited by Stan Pope on Sun Dec 05, 2010 10:21 am, edited 1 time in total.
Reason: Add missing "that" in "Each racer is eliminated..."


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

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: N-elim format lane selection

Post by Stan Pope » Sun Dec 05, 2010 10:49 am

Stan Pope wrote:From the simulations, you can see how many Cubs leave without winning a race against their peers. The number is not very large. If you want everyone to leave with a contested win (i.e. excluding bye wins), ...
Argh!!! The simulation does not compute well for 2 lane tracks to show the end-of-group bye run count. For instance, the 11-car simulation first round has 5 contested heats and one bye. That uncontested heat did not get removed from the "Total Contested Runs" count. There are 10 more uncontested runs than the simulation shows. (Look for odd numbers before the group size reduces to 1.)

I think it was a case of selective blindness. I have always run this method on 3-lane tracks, and with 3 lanes, it is easy to keep all runs contested until the group size is reduced to one.


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

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: N-elim format lane selection

Post by Stan Pope » Sun Dec 05, 2010 6:52 pm

Stan Pope wrote:
Stan Pope wrote:From the simulations, you can see how many Cubs leave without winning a race against their peers. The number is not very large. If you want everyone to leave with a contested win (i.e. excluding bye wins), ...
Argh!!! The simulation does not compute well for 2 lane tracks to show the end-of-group bye run count. For instance, the 11-car simulation first round has 5 contested heats and one bye. That uncontested heat did not get removed from the "Total Contested Runs" count. There are 10 more uncontested runs than the simulation shows. (Look for odd numbers before the group size reduces to 1.)

I think it was a case of selective blindness. I have always run this method on 3-lane tracks, and with 3 lanes, it is easy to keep all runs contested until the group size is reduced to one.
That error has been corrected. Hopefully, no new errors have been added in the process. :(


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

dfscott
Journeyman
Journeyman
Posts: 40
Joined: Fri Dec 03, 2010 1:34 pm
Location: Atlanta, GA

Re: N-elim format lane selection

Post by dfscott » Sun Dec 05, 2010 7:46 pm

No worries and thanks for all the help! I'm going to give this a shot and see how it goes. But regardless of the result, the amount of mistakes that this web site has saved me from making has been well worth all the time I've spent reworking my race plan.



Post Reply