60 boys, 3 lane track w/ timer, New PWD Coordinator

[Warthog]

Well I am just getting into this scheduling of a Derby Race and have looked through the posts and I believe this is a new issue to bring up or at least more specific to our race...

In taking over the PWD Coordinator role, I am not that familiar with the different scheduling methods but I hope to take some things that I liked from our Pack's previous races as well as make some changes and I would like some input. I am hoping to get approval to order the GPRM software to aid in the scheduling this year.

Our previous Pack races have separated out each den to race against each other. They bring each den up along side the track to watch their cars race, one den at a time. When all dens are done they award places within each den as well as best places in the Pack based on average time.

I like the dens being broken into groups, but I would like to see a final runoff of maybe the top 6 cars based on their average times from the den races. With a 3 lane track I would like to have each car run twice in each lane for both the den races and the run-off race. Any thoughts, suggestions?

Thanks

[Stan Pope]

GPRM includes PPN schedules, which assure lane equity, with one or two runs per lane scheduled. It also, provides sufficient variety of opponents that comparative heat finish places should substantiate the timed results. (Simple rotation schedules, e.g. 1-2-3, 2-3-4, 3-4-5, ..., can provide accurate timed results, but the heat places can't be used to substantiate timed results.)

The possibility for track changes during the day justifies running a final. A good bump on the track can cause later cars to run slower or faster than before the bump!

If your track's starting gate is totally manual, i.e. the speed of movement of the gate is determined by the operator, then you may find unintended variances in heat times. I've seen it around here. But if the gate is spring-loaded, or if you can get the pack to modify the gate to make it so, then timed competition within the den's should be solid.

Double check on what results your races may have to produce. If your district/council races are based on pack qualification, then check on whether you need top few in pack or top few in age groups.

Another thought ... look at possibility of boys racing their own cars rather than watching adults race them. It requires extra work, but IMO the extra work is worth it to the boys. Viability depends heavily on area traditions.

[Warthog]

Thanks Stan, I have always enjoyed and appreciated reading your input on this forum.

If your track's starting gate is totally manual, i.e. the speed of movement of the gate is determined by the operator, then you may find unintended variances in heat times. I've seen it around here. But if the gate is spring-loaded, or if you can get the pack to modify the gate to make it so, then timed competition within the den's should be solid.

This has been an issue at the previous races. The starting gate has been totally manual - which has lead to some very inconsistent times. I am going to install a spring loaded start gate to resolve this issue.

Double check on what results your races may have to produce. If your district/council races are based on pack qualification, then check on whether you need top few in pack or top few in age groups.

The top 3 finishers in the Pack move on to the District Race.

Another thought ... look at possibility of boys racing their own cars rather than watching adults race them. It requires extra work, but IMO the extra work is worth it to the boys. Viability depends heavily on area traditions.

The local Boy Scout Troop has run our race in the past as far as running the cars. Having the boys run their own cars is something to consider.

Thanks

[Stan Pope]

The top 3 finishers in the Pack move on to the District Race.

Then how would you identify those top 3? Possibilities:

1. Top 3 from the finals involving den winners? Often, the second or third fastest car in the pack loses to the fastest car in the pack during den racing! Then your pack's second or third fastest car would be excluded from the district races!

2. Best 3 times or average times from Den racing? A good bump of the track can affect the times of cars following, making time comparisons unreliable. Then your pack's fastest cars might be excluded.

Is there anything that you can do to validate the track's consistency throughout the day? For instance, if den winner's times during den racing and during finals were unchanged, then you would have confidence that the track stayed unchanged through the day. That won't happen of course ... there will be variances! But, how little variance in den race and finals times is needed to provide the needed conficence? Time to call in some high-powered statisticians, maybe?

[Warthog]

I had seen something similar to this in another post where another Pack ran their races by Den and then took the top 3 cars from each Den to be in the final round. To me this could be unfair if say one Den had 5 of the fastest cars in the Pack.

I would like to take the cars with the fastest time averages from all the Den races regardless of which Den they come from. I am just not sure how many to take to the finals. Six seems a good number to mainly determine the top 3, but I could also see taking the top 9. The only thing with bringing more cars into the final round the longer the race will last.

[Warthog]

Is there anything that you can do to validate the track's consistency throughout the day? For instance, if den winner's times during den racing and during finals were unchanged, then you would have confidence that the track stayed unchanged through the day. That won't happen of course ... there will be variances! But, how little variance in den race and finals times is needed to provide the needed conficence? Time to call in some high-powered statisticians, maybe?

Well, somethings that can be done during the race or between the Den Races and the finals to ensure the track is in consistent condition:

-Secure track - use a yellow tapes or flags to keep scouts back from track to aid in keeping the track from being bumped.

-Clean track surface between Den Races and Finals.

-Ensure track joints are smooth, and that track is level in two dimensions.

Other than that variances between den races and finals could be caused by other events unrelated to the track - wheel becomes loose, lube wears out, axle bent, particles in bore, etc.

Overall, most, if not all cars should show little variance. If there is a significant variance by all the cars, that would indicate something changed with the track.

[Cory]

Is there anything that you can do to validate the track's consistency throughout the day?

You could do a control run before and after every heat.

Just curious, what do folks in the "times are better than points" camp do at the end of the day to validate track consistency? And what is done when it is discovered that the track was inconsistent?

[rdeis]

In the two district races I researched last year, they didn't.

They assumed that time variances from run to run do not exist, so all they had to worry about is lane bias.

Lane bias is easily overcome by running on each lane and averaging, but other sources of error are ignored.

[Mr. Slick]

Well, I check the variance for each car to make sure it is "reasonable for the track". This is usually less than 3% for any specific car on the 64' track. If there are cars that have a variance of greater than 10% from their average time, they are checked out. If it was not due to a car change/failure, the car gets to race down the lane again to see if it can get a time that is more consistent with it's other times. In a group of 40 cars this may happen to one or two cars. They usually are the ones with a wide range to start with or are showing a consistent slowing/speeding of results.

When we help out a slow car, usually determined after a race or two, the times will change lots and so notes of what cars are "fixed" during the races are used. If the times of the car are not in the trophy range I encourage the scout to re-race the slow ones to help in the overall standings. Raceday fixes can really help a car but rarely will they take a slow car and make it into a trophy winner. I have helped adjust cars that were real slow and made them trophy class AFTER the races are over to help a cub that is really down in the dumps about his car. This usually takes a pair of axles and some lube with several test runs to get it aligned.

Otherwise I will have to say that if the distribution of the cars through out the races is consistent than all of the cars have the same probability of having a race after any specific time altering event.

The "track events" that are the worst are the debris left on the track by a previous car. . . sometimes not caught for a few races -- observant parent notices a consistent "new bump" in the track. The severity will dictate the response. . . large changes will show up in the maximum variance from the average for the car.

[Cory]

Otherwise I will have to say that if the distribution of the cars through out the races is consistent than all of the cars have the same probability of having a race after any specific time altering event.

Thanks for the explanation.

Have you been able to analyze your data to discover "time altering events" of the type quoted above -- i.e. events that would affect all cars across all lanes? I've heard anecdotal tales of lane-rotation scheduled, time-scored Derbies where it has been argued that such events have occurred.

E.g. a car races in lanes 3 & 4 of in the first two overall heats, then later returns to race in lanes 1 & 2 in the last two overall heats -- times for the first two runs are consistent, as are the times for the last two runs, but the two pairs of runs are far from consistent, time-wise.

But since the "analysis" involved only one car it's not valid to blame the track for the inconsistency. A more rigorous study which considers a larger sample of cars is required.

[Mr. Slick]

Well, I have lots of race data. . . how would you like it? www.PwdRacing.com has several races from last season posted and several years worth of district results also.

Maybe we can come up with a method of doing the data archiving to aid in the analysis of the numbers for the upcoming season. Maybe GPRM will have an add-on to aid us.

I have a couple dozen races scheduled already so I should have well over 1000 cars worth of data for the next season. I frequently see cars more than once. Sometimes a car shows up 3 times, at a pack race and then again at the district races and in the sibling division in the Girl Scout Service Unit(BSA district equiv).

[Cory]

Well, I have lots of race data. . . how would you like it? www.PwdRacing.com has several races from last season posted and several years worth of district results also.

Thanks, I'll take a look. Not sure at all how to start analyzing it, though....