Scheduling Methods thats needs help

[Scoremaker]

Okay Ladies and Gentlemen I need help on this.Last year my son went to race in the Council race.There where 13 cars in his divison.They had a 8 lane track each car was going to race one time on each lane.To get there total time so they can put them into a bracket system.Then they where going to use only the 2 oustside lanes.This where I need your help at.How would you place the cars in the bracket system.Would you place them how they place in the time trails or would you mixs them up so you can have the top four cars racing in the final four.And who would you give the bye to and where would you place that car in the bracket system.I have more to say about this Scheduling Method.

[Stan Pope]

Okay Ladies and Gentlemen I need help on this.Last year my son went to race in the Council race.There where 13 cars in his divison.They had a 8 lane track each car was going to race one time on each lane.To get there total time so they can put them into a bracket system.Then they where going to use only the 2 oustside lanes.This where I need your help at.How would you place the cars in the bracket system.Would you place them how they place in the time trails or would you mixs them up so you can have the top four cars racing in the final four.And who would you give the bye to and where would you place that car in the bracket system.I have more to say about this Scheduling Method.

From the preliminary timings, you have a rough ranking of cars, fastest to slowest. If the next round is to have the cars divided into 4 brackets, then I would distribute them following this pattern:

1->A, 2->B, 3-> C, 4->D, 5->D, 6->C, 7->B, 8->A
Repeat the ABCDDCBA scattering for each of the following groups of 8.
The reason for the reversal for the last half of each group is to give the slight advantage of facing car 8 rather than car 5 to the car which earned it. (It also eliminates some potential gamesmanship.)

At this point, there are several possibilities, depending on how many cars are in each group and what kind of finish line capability you have.

If you can reliably select heat place finishes (e.g. which car finished 1st, which was 2nd, which was 3rd, etc) then I'd go with a 3 or 4 lane PPN and promote the 4 highest scoring to the finals.

If you can only reliably select heat winners, then I'd go with quadruple elimination and promote the 4 "survivors" to the finals. If you have about 28 or more in each bracket race 3 at a time. Fewer, race 2 at a time.

For finals you would have 16 cars racing. I'd run PPN on 4 lanes, select 7 highest scoring and run PN or CPN on 3 lanes to make final trophy determination.

That is the answer I'd give IF I had to abide your preconditions.

If I could "start the design from scratch," I'd forget about that unwieldy 8-lane track and go with a good 3 or 4 lane Piantidosi track.

If the number of participants competing against each other were about 50 or less, I'd use a PPN screening / PH-7 finals method.

If more, I'd run a quintuple elimination, racing 3 at a time. If I had more than about 80 Scouts, I'd break into brackets until the number was down to about 60 in each bracket and promote 5 survivors from each bracket to final.

Note: I'd only use quadruple or quintuple elimination format if the boys staged their own cars. They participate in the sequencing and opponent determination / randomization. Otherwise, there can be an appearance of manipulation and/or arbitrary decisions in pairing the cars for each heat.

More details on these methods are available on my website.

[Scoremaker]

Okay Stan you need to stay with this program not going off by another system or program.What im trying to say when the cars get done doing there time trails on the eight lane track they are place into bracket a system.There are ONLY 13 CARS?I know there's and odd numbers of cars in this field.I wanted to know who would you give the bye too?THEN I SAID IM ONLY USING THE TWO OUTSIDE LANES ONLY?The cars are running a best out of three race.Then I ask how would you place each car into a brackets system by the way they place in the time trails or mix them up.If I had a say I would do it like the NCAA Basketball bracket are played by Rank 1vs16,8vs9,5vs12,4vs13,6vs11,3vs14,7vs10,2vs15.

[Stan Pope]

There are ONLY 13 CARS?I know there's and odd numbers of cars in this field.I wanted to know who would you give the bye too?THEN I SAID IM ONLY USING THE TWO OUTSIDE LANES ONLY?The cars are running a best out of three race.Then I ask how would you place each car into a brackets system by the way they place in the time trails or mix them up.If I had a say I would do it like the NCAA Basketball bracket are played by Rank 1vs16,8vs9,5vs12,4vs13,6vs11,3vs14,7vs10,2vs15.

You don't say how the competition is to be conducted among the 13 cars in the bracket, except that each "heat" is best 2 out of three runs using the outside lanes.

The competition within the bracket could be done with a full round-robin or, even double round-robin, although that would negate the need and desirability for doing best 2 out of three runs at each "heat".

The chart for those round robins can be generated at http://members.aol.com/standcmr/ppngen.html using values of lanes: 2, cars: 13, and rounds: 6 or 12.

Once you see the pairings, rerun the chart with heat ordering options "medium" for all three criteria. That resequences the pairings to keep heat counts nearly even as you progress through the heats. and avoids the same car repeating in successive heats (to speed things up a bit.)

If you select the 12-round chart, then every car will have raced against each other car twice and in swapped lanes.

I suspect what is planned for the "bracket" is some kind of "elimination scheme", and your question is who should race against whom in the brackets.

I would pad out with three byes (ranked number 14, 15, and 16) and go with a simple charted single or double elimination, laid out in a tree as you described (1vs16,8vs9,5vs12,4vs13,6vs11,3vs14,7vs10,2vs15). This gets rid of the byes quickly and pairs the byes with the three fastest cars. If a double elimination chart isused, check to assure that the byes do not race each other in the 2nd chance bracket. The 16-car DE chart on my site would satisfy that requirement. This delays the meetings of the probably faster cars until late in the competition, which can be an advantage in interpreting results.

Are there any more stipulations that should be factored in?

[Stan Pope]

I omitted the address for the DE chart... it is at
http://members.aol.com/standcmr/dechart.html

The cars are running a best out of three race.

This is a questionable practice. A lot of time is used with little improvement in accuracy of results. If the difference between the cars exceeds the difference between the lanes then one car sweeps, otherwise, the result depends on which car gets the faster lane in the third and deciding run.

The practice does allow a car to sometimes come back after a bad run. However, good cars don't have bad runs unless there is a severe track defect or the car is staged badly.

Some pitfalls include:

Arbitrary assignment of cars to lanes for the first heat and alternating thereafter. This means that someone is deciding which car runs in which lane; in effect, deciding which car gets the advantage from lane differences. The lane advantage should be done by lot.

Stopping the "heat" after two runs if one car wins both. This gives the winner some advantage frem fewer runs (less graphite shake-out.)

It sounds like you are saddled with some traditions. Those can be hard to improve upon. Folks want to do it that way because "That's the way we've always done it."

We successfully broke a long tradition of running double eliminations for each age group and giving trophies to the 5 top cars by "trying" a quintuple elimination on the youngest age group. (Needless to say, the third through fifth place trophies involved a lot of luck in the initial chart placement.) The try was a smashing success from several points of view, not the least of which were more runs for each scout, fewer byes, less work for the chairman, and more racing per minute. The next year all the age groups went quintuple elim! Of course, we were running 50 to 100 scouts in each age group, so our method choices were somewhat limited.

[Scoremaker]

I for got to put this in last time when they race on the outside lanes they change lane each time.Its Double elimination format.

[Stan Pope]

While it is outside of your parameters, I would feel remiss if I didn't mention that you can conduct a full round-robin for 13 cars racing 4 at a time in 13 heats and a double round-robin for 13 cars racing 4 at a time in 26 heats. The double round-robin has each pairing race in alternate lanes. Using the chart generator mentioned earlier: 4 lanes, 13 cars and 1 or 2 rounds.

This is a final standings method based on accumulated points. Accuracy is top-notch. Advantages include time efficiency and the fact that it becomes rather clear when lane differences exceed car differences, yielding, at the end, trophy place ties that must be broken.

Too bad you are bound to those poor traditions.

[Scoremaker]

Okay Stan I still think you do not get yet?THE PACK THATS RUNS THE COUNCIL RACE ONLY USE THE TWO OUTSIDE LANES ONLY.Its a best out of three race the car change lanes each time they race Only TWO CARS ARE RACING.Then I ask who would you give the BYE to and where would you place them in the WINNERS OR LOSSER Bracket Then I said the NCAA Basketball Tournament system work really well by Rank.When the cars are place into the bracket THIS IS A DOUBLE ELIMINATION ROUND IN COUNCIL THEY ONLY PLACE THREE POSTION ONLY.They have a eight lane track you races one time on each lane to get your total time.BY USING THERE FORMAT NOT ANY OTHER FORMAT JUST THIS FORMAT.HELP HELP HELP.

[Stan Pope]

I answered that question at 1:30 CDST this afternoon:

I would pad out with three byes (ranked number 14, 15, and 16) and go with a simple charted single or double elimination, laid out in a tree as you described (1vs16,8vs9,5vs12,4vs13,6vs11,3vs14,7vs10,2vs15). This gets rid of the byes quickly and pairs the byes with the three fastest cars. If a double elimination chart isused, check to assure that the byes do not race each other in the 2nd chance bracket. The 16-car DE chart on my site would satisfy that requirement. This delays the meetings of the probably faster cars until late in the competition, which can be an advantage in interpreting results.

I gave you the URL for that DE chart also.

I would not try to use a "13-car" double elim bracket because I think that it would not keep graphite shakeout equal.

I think that this addresses your question exactly. If not, please clarify.

[Stan Pope]

I would pad out with three byes (ranked number 14, 15, and 16) and go with a simple charted single or double elimination, laid out in a tree as you described (1vs16,8vs9,5vs12,4vs13,6vs11,3vs14,7vs10,2vs15).

This sentence answered your question of which racers would bet the bye. It occurred to me that the answer is somewhat indirect, so you might have missed it this afternoon. Please note that if the bye(s) are ranked 14, 15 and 16 as I suggested, then they would race against 1, 2 and 3 in the first round.

See the chart at http://members.aol.com/standcmr/dechart.html to see where I where I would place the ranks in the initial round. It all follows from there.

You spoke as though they propose to use a 14 car DE chart. I am aware of none that would be clean, but if you show me the picture, perhaps I can understand your problem better.

As for the other stuff ... you deserve to know that there are better solutions than what the pack that runs the Council Derby is trying to pass off. You may not be able to use the information this year, but maybe next year or the year after.

[Scoremaker]

Thanks Stan I understand your Method.When my son race at Council I told him thats the best of the best would be racing here.It's how they postion the cars in the brackets after they did there time trails.1st played 2nd and so on down to 13.All the parent's thought the top 6 cars did not even had a chance to place's the bottom seven had a better chance's.All I know my son had a good time racing and making it that far for his 1st year as a Tiger.He's District Champion and Place fourth over all in Council.If this Pack runs this race next year I will ask them to use this method so everyone has a far chance,and see the top three cars go at it.

[Stan Pope]

Out of curiosity, how did your son's rank in the timed races match up with his final rank?

You said that your son's division consisted of 13 Scouts. Where these the only ones competing against your son?

What were the overall numbers and breakdown for competition? For instance our district will have 300+ Scouts participate in the District Derby. They compete by age groups (5 groups, Tiger to Webelos II) for 1st through 5th place torphies. These races are quintuple elimination. The 1st place racer from each age group compete for the grand championship trophy. This has been done in a variety of ways ... currently we use a 5-car PN chart on a 4-lane track.

Back to your situation, you might head off the problem by communicating with the Council Committee which makes the decision re organizing the Council Derby. If you can (1) show the nature of the inaccuracy of the method used, quantified if possible, and (2) show how accuracy can be improved, they might listen.

On the other hand, they might not ... traditions are sometimes very strong. Shortly after my grandson swept the council races over in his council, I wrote a critique of their racing methodology and shipped it off to the council office along with a CD full of pictures of the event. We attended again last year and I saw little effect from my efforts. They did make all four of the preliminary's tracks similar (instead of the one long aluminum track and 3 shorter Piantidosi tracks). They still took the two DE survivors from each preliminary track and raced them in an 8-car DE chart to award 4 trophies. (I bet that the 3rd or 4th fastest car entered at the races did not receive a trophy.) Anyway, the solution to that problem is to go in with the fastest car. Then it is harder for them to mess up your chances. They can still do it, of course, but it is harder for them!

[Scoremaker]

My Son rank in his time was third.There where 13 Tigers Competing against my son and maybe 50 or 60 Wolfs,Bears,Weblos 1and 2.They race togther to get the overall winner in that group.They also do the same at the District race level Tigers race against Tigers and Wolfs Bears Weblos 1and 2,but they race everyone all at once and do a time trail with a three lane track.Our Pack runs the District race there's at lease 300+ scouts racing that day.And how can I find out who is doing the Council race for 2004?Do I call the Scout office?So I can send them some info,and if the same Pack that run the Council race for 2004 will understand what every Scout went thru and parent.

[Stan Pope]

My Son rank in his time was third.There where 13 Tigers Competing against my son and maybe 50 or 60 Wolfs,Bears,Weblos 1and 2.They race togther to get the overall winner in that group.They also do the same at the District race level Tigers race against Tigers and Wolfs Bears Weblos 1and 2,but they race everyone all at once and do a time trail with a three lane track.Our Pack runs the District race there's at lease 300+ scouts racing that day.And how can I find out who is doing the Council race for 2004?Do I call the Scout office?So I can send them some info,and if the same Pack that run the Council race for 2004 will understand what every Scout went thru and parent.

That is about as much accuracy as you can expect from a DE chart if cars are seeded without regard to their speed. Accuracy suffers when comparable cars race together early, and improves if they are seeded according to my DE chart with "1" being fastest, etc. However, a DE chart, in any case, does a poor job for 3rd and 4th place, and often a worthy racer is denied a 3rd or 4th place trophy.

I think that when a serious improvement suggestion is filed, it should get to two places: the person or group who would / should perform differently, and the person or group that oversees that group. A single communication, if you expect it to reach both, should go to the overseer, in this case, a Council Committee. Your District Executive should be able to find out the name of the chairman of that committee. Then address accordingly, c/o your BSA Service Center.

Is your district race run strictly on time? If so, do they average all runs? or discard some high and low runs? Detail! Is there other racing other than timed? If so, how is it conducted? What do they do to "prove" the electronic timings? (Does someone certify the clocks? If so, how?)

[Scoremaker]

At the District Level they average all the times up.It's strictly time,and all time count.There's one man at the starting gate that start the time and one at the lap top to watch the clock,and one at the end of the track.Do you suggest that I contact the Council Committee and the District Executive now?Council race doesn't start until May.