Advancing to finals - how many do you do??

[jeffd]

In the past our cub scout pack has advanced the top (tiger - cub - webelos) entry to a pack finals race to determine the pack speed champion.

We now have a timer with our track, and with our software (GPRM 7.0) i now have the option to make the finals round with more cars. To me, it would seems as if advancing - lets say - 4 cars per group to the finals would allow more chances for a car to squeak out a win.

So what would you say?

1) Advance the top car only from the racing heat and get top standings from there?

2) Advance 4 (or some other number) of cars to a finals round where more than 1 car would have a chance to squeak out a win?

3) Something else?

[pack529holycross]

In the past our cub scout pack has advanced the top (tiger - cub - webelos) entry to a pack finals race to determine the pack speed champion.

We now have a timer with our track, and with our software (GPRM 7.0) i now have the option to make the finals round with more cars. To me, it would seems as if advancing - lets say - 4 cars per group to the finals would allow more chances for a car to squeak out a win.

So what would you say?

1) Advance the top car only from the racing heat and get top standings from there?

2) Advance 4 (or some other number) of cars to a finals round where more than 1 car would have a chance to squeak out a win?

3) Something else?

We have 58 boys registered as Scouts. We opened up to siblings and parents due to the advance to a 6lane track this year. We ended up with 48 scouts, 14 siblings, and 7 parents. We ran 6 rounds ( one per den ), and advanced 3 out of every den to the finals series of 8 racers. In the end 30 kids got to race 6 times each, and 18 kids got to race 12 times each. We ran the siblings in their own group and gave trophies for top three and certificates for the remaining racers ( awards like " most stickers " or " most likely to get a speeding ticket " ).

[Stan Pope]

In the past our cub scout pack has advanced the top (tiger - cub - webelos) entry to a pack finals race to determine the pack speed champion.

We now have a timer with our track, and with our software (GPRM 7.0) i now have the option to make the finals round with more cars. To me, it would seems as if advancing - lets say - 4 cars per group to the finals would allow more chances for a car to squeak out a win.

So what would you say?

1) Advance the top car only from the racing heat and get top standings from there?

2) Advance 4 (or some other number) of cars to a finals round where more than 1 car would have a chance to squeak out a win?

3) Something else?

First consideration I'd mull is, "What would be different in the finals than during the grade races?" It could be that
1. We suspect that the racing environment might not be stable.
2. We want to include lubrication durability in the criteria for "fastest" (assumes no relubrication between grade races and finals).
3. We suspect that inherent variability in each car from un to run (even if in the same lane) limits the accuracy of the grade race rankings.

If these factors are absent (or reasonably believed to be negligible), then there is really no reason to run a finals, because the results should not change, We might run 'em "for the excitement", but we expect the times posted in the grade races to hold up.

If we suspect that the racing environment might have changed through the day, then we should run finals.

If we suspect that inherent variability of cars affected grade race accuracy, then we should include a few more cars from each group in the finals. (Inherent variability affects accuracy when the variablity within each car exceeds 1/2 the variability between cars. As the ratio of those variabilities increases, accuracy in ranking decreases.)

If we want to include durability in the definition of fastest, then we might well want to include a lot more cars from each group in the finals.

[ohiofitter]

Our troop has the same format top 5 cars run for the pack overall...But our troop has only a short period of time to use the school cafeteria...So they only run the five...But time allowing I would rather see more cars run for the title sort of speak

[Stan Pope]

Our troop has the same format top 5 cars run for the pack overall...But our troop has only a short period of time to use the school cafeteria...So they only run the five...But time allowing I would rather see more cars run for the title sort of speak

Yes, I understand!

There is a "minimum time solution" to such problems. But, we don't have to use that one if spending more time provides an equally correct solution and provides more activity for the participants. So, generally, I think in terms of "What is just enough to do the job correctly? Now, what can I add in the time available to make it more fun yet still do the job correctly?

But, sometimes you have to just do the best job you can in the insufficient time available.

[3 Cub Dad]

At the risk of re-starting the Points vs Time debate, (which is covered in multiple strings!!!) I'll tell you what we did last year, this year, and am planning for Districts.

2 years ago, the pack used double elimination, awful. The fact is, the kids want to see their cars race. The longer you can keep the majority of the kids cars involved, the more involved they are! Sitting watching other Dens run also isn't that "involved" for them. We have a relatively small pack, a little over 30 kids. (two of the Dens have only 3 or 4 kids)

We run everyone at the same time, using the random perfect-N chart generator on GPRM. We have a four lane track and every car races in every lane twice. So every kid gets to see his car run at least 8 times. We run these preliminaries using points. (more to come on this) We don't have a pre-set number to advance to the finals. We let the results of these runs dictate how many to advance. Last year, the top 7 cars were close in points. So they were advanced regardless of Den. Another perfect-N chart and they raced for championship by points. Trophies for top four positions.

Now to discuss the results, (having the data on GPRM you can go back and manipulate the data to see how they would have ranked by avg time, total time, etc!) 6 of the top 7 cars were from the Wolf pack, (who all came to our workshop in my garage!). Would it have been fair to have a Tiger car or Bear car that was nowhere near their speed to "race" for the championship? Wouldn't have been much of a race or been that exciting.

In comparing the results: The top 7 cars in points were also the top 7 in avg time, only the "positions" change. I believe that the points system gives you a great way of determining the "group" of fastest cars, and then they can race off head to head. We did have one track issue halfway through the Derby, IF we had been running times, this could have been controversial. Also, we had the starter accidently reset the timer on one race so it didn't record the time. We didn't have to re-run because the display still called the finish order. All in all, all kids eight races, plus another 7 races final, in under 2 hours.

Depending on the number of cars in the District this year, (we haven't had one in 5 years) I may advance to finals off of avg time after each car runs once in each lane. (oh yeah, review of the data from last years race shows no statistical variance or bias between lanes)

Sorry this is long, but hope it helps.

[jeffd]

Thanks for the responses.

This is my second year in the pack and i dont think anyone knew the real method behind it all... we just knew cars raced - some won - some lost - and it was up to Jim to say what cars raced

[speedbump]

What would you suggest for a small pack (under 30)? We have always ran double elimination per den and advanced the den winners to the "pack finals" like stated above. I need a better alternative. I feel we are eliminating too many fast cars via this method. We can send 5 cars to District and I would like us to be well represented. Unfortunately, we do not have a timer just light triggered finishes (4-lane) I need help......Is it that difficult to run quad-elims or some sort for the entire pack?

[3 Cub Dad]

What would you suggest for a small pack (under 30)? We have always ran double elimination per den and advanced the den winners to the "pack finals" like stated above. I need a better alternative. I feel we are eliminating too many fast cars via this method. We can send 5 cars to District and I would like us to be well represented. Unfortunately, we do not have a timer just light triggered finishes (4-lane) I need help......Is it that difficult to run quad-elims or some sort for the entire pack?

With the four lane track, without a timer, I think the best method would be a perfect-N chart and run by points. Stans site has some great info on the charts, but in my opinion, (worth what you're paying for it!) I'd get a copy of Grand Prix Race Management software, which will run the schedules for you. If you can't plug the light trigger into the computer, no problem, you just enter the finish order after each run, (very simple). Running by points is basically 1 point for first, 2 for second, 3 for 3rd, 4 for 4th and 5 for DNF. The schedule can make sure that every car races once or twice in each lane, and races against all of the other racers. You can then either award from these results, or advance a small group to the finals and have anywhere from 4-8 race for the championship. In this manner, the majority of the kids are involved in the racing for 80% of the Derby. Let's face it, the kids want and need to see their cars running on the track. A couple of trips down and out, (double elimination), is in my opinion the worst method out there. A lot of people have strong feelings of "points vs time", but without a timer the point is moot. If you send Stan a PM, he can direct you to some of the info.

Hope this helps.

[Stan Pope]

We can send 5 cars to District and I would like us to be well represented. Unfortunately, we do not have a timer just light triggered finishes (4-lane) I need help......Is it that difficult to run quad-elims or some sort for the entire pack?

Quadruple (4) or Quintuple (5) Elimination is really easy and fast to run. They are especially effective when the boys stage and retrieve their own cars. If that is something you would like to look at in more detail, get with me via email and I'll help you get a plan together.

A "whole pack" points competition also works. It an be run on paper or on computer. For this, I'd avance exactly 13 to the finals, since this will improve the accuracy of identifying the fastest 5 in the pack. I've run paper (actually record keeping on transparencies projected on a screen) and that works pretty well, too. It is a bit slower if the boys race their own cars, since invariably one of the boys in each heat is daydreaming and not on deck ready to stage his car. But it works, too. Same offer on this. In fact you may wish to look at both alternatives, get plans together and get input from your pack's leadership to make a final choice.

If your representatives are supposed to be the 5 fastest cars in the pack, then that is how your pack competition should be geared, and den races are not appropriate if they are part of the selection process... unless the den races promote the fastest 5 in the den to finals!

[speedbump]

Thanks for the responses. Just trying to dig up a few different ideas to pitch before a final decision is made. The pack has always run double elims per Den and transfered each Den winner to the Pack finals (also double elim). I don't mind it really, but the size of the dens vary so much, I feel that some of the better prepared cars are being left out because of this. For example, my son's den has 9 kids and most of the other dens have 2-4. That eliminates 8 kids from having a shot at the Pack trophy whereas a kid from the other den of TWO gets in (in my eyes) by default because he only had to beat ONE car or even worse if he is the only kid to show up from his den. Hope you can see the problem. Not bragging but my son has won his den and pack race the last 2 of 3 years and came in second the year he did not win. So, I am not here to complain ...and since we come from the heaviest pack of 9, a lot of good competition is being shut out....
I will pitch the idea of trying the PPN. I like the idea of it and helps keep the kids interested throughout the race time. 25 kids on 2 lanes using 2 rounds alternating lanes = 50 heats. Then transfer the top 5 in points to a PACK shutout or something...Just mulling over some ideas.....Thanks again

Stan, curious, how complicated is the quad-elim? I understand the concept, just have never seen it implemented

[Stan Pope]

Stan, curious, how complicated is the quad-elim? I understand the concept, just have never seen it implemented

If the boys race their own cars, i.e. bring the from the pit to the starting line, then it is easy.

The tricky part is randomizing the opponents and that is easily and very transparently accomplished, too.

If the boys are spectators, then you have to arrange a separate way to assign opponents for each round. One of the guys posted his method last year here on DT.

The basic concept is that racers with like records race each other until each score group has one member left. Then run a ladder finals.

Each racer races in each "round" until he accumulates enough non-wins to retire. The progression is from group 0 to group 1 to group 2 ... Winning his heat keeps him in the same group, otherwise he moves on to the next group.

Each group starts with the group with the most non-wins. The first round starts with everyone in group 0. When the round is complete, there are racers in groups 0 and 1. So start round 2 with group 1, racing to form a new group 1 and a new group 2. The goe to group 0, racing them to form a new group 0 and add to group 1.

To randomize, I usually let the boys in the group line up as they wish, then the draw for lanes sorts them out and rearranges them. For a two lane track, they may blind draw from a container with tokens that indicate "Race in lane 1" or "Race in lane 2" or "Get back in line" or "Go to the back of the line" (one or more of these, depending on how much randomizing you need or can put up with.) As you approach the end of each group in each round, there may be fewer racers than tokens to draw from Then leave the "Race" token in but remove excess "Get.." or "Go" tokens.

Poker chips (Red White and Blue colors are cheap and easy to use for this purpose. As each pairing is determined exchange the "race" token for distinctively colored numbered chip as their ticket to stage their car.

The Judge at the finish line needs to direct boys to their proper group, same # or next higher # depending on wins or not. If judges can't agree on winner, treat 'em both as winners for the heat!

Email for more if you need.

[derbyspeed]

We only have 23 scouts and that's if they all race. We are doing point this year with a preliminary and a finals. We will give away 4 trophies. How many should we advance to the finals - 8? Using the gprm software.

By the way, we do ours as a full pack race and then award medals to the scout in each Den who goes the farthest in the race for his specific den. This eliminates having slow cars in the finals but rewards each den. So even if they don't make it to the finals they still get recognized.

But I am still a little confused on how many to send to the finals, any help is appreciated.

Mike Webb
http://www.cubscoutpack118.com/1.html

[jeffd]

Mike – though our pack has more scouts than yours, I ended up picking a number based on the number of lanes we have on our track. This way I ensure an even number of races with a full lineup (no open lanes).

In GPRM we advance the top 4 from each group (Tiger, Cub, Webelos).

[3 Cub Dad]

Mike,

I know this may sound a little weird, but especially when we're using GPRM, we don't pre-difine the number for the finals, (we use points too). We look at the results and see if there is a natural break. Lets say there were 20 racers, and when preliminaries are done, four are real close in points, advance those four, 7 real close? advance those seven. GPRM makes the chart real quick. Last year we advanced 7. I remember reading that 7-8 with a four lane track was a good amount for the perfect-n chart generator