PPN Charts / Sterns Charts ?

[DMWOOD]

We have a 3 lane track and most of the PPN charts are for 2 rounds (6 races ea.) once you get above 25 racers. I would like to get some charts for the 25 - 40 racer range that allow for about 3 PPN rounds (9 races ea.). I want the kids to be able to race as much as possible for the work they put into the cars.

Is there a good way to add the extra round to the PPN charts?
OR
Is there a good Sterns Chart generator that I could use to generat these charts?

Thanks for any help you can provide.

[Stan Pope]

You can take a couple of tacks on this.

Knowing the following:
1. foibles of Stearns,
2. you will be running PPN charts, so the opponent balance is as good as you can get, but not perfect, and
3. the slower cars probably don't want to be beaten for the 7th, 8th and 9th times,
I'd opt to run a 2-round ppn involving all of the scouts and a CPN finals ... (2-round 7 car final or 4 round 13 car final) to decide trophies. These charts are perfectly balanced and will produce excellent accuracy. A PN finals (1-round 7 car final or 2 round 13 car final) will also do a good job for finals, but may be slightly less accurate than the CPN chart.

Alternatively, accept that the lane balance and opponent balance will be lacking and go with Stearns. Be sure to have the boys draw for racing numbers so that any inequities will be obviously allocated by chance!

If you opt for Stearns, you can still take the separate finals approach as with PPN ... select about twice as many finalists as you have trophies to award (or places to be determined) then push that number to the best choice between 7 and 13, and run them on a PN or CPN chart.

[Stan Pope]

Another alternative is to run a second one or two round chart with the scouts redrawing race numbers. This may result in greater opponent inequity, but if the numbers are drawn rather than assigned, at least it is fair. PPN guarantees max opponent inequity of 1. Running two charts as suggested would guarantee max opponent inequity of 2 (I think ... I'm running this extrapolation on intuituion.)

The obvious problem is tracking the racer's two numbers correctly and adding their scores. But I would not just run the same chart with the same race number assignments ... those issues have already been decided!

[Cory]

Is there a good way to add the extra round to the PPN charts?

How fast a computer do you have?

The charts that are available in the PPN generator are the ones that we know of at this time.

It's easy to create a CPN chart from a PN chart. However, there is no known algorithm for coming up with the various rounds of PN and PPN charts. With no algorithm, this means that one must hunt for them by checking the myriad permutations which are available. And while computers are good at this sort of hunting, if you're talking about large numbers of cars/lanes/rounds, then the problem quickly gets much too big for even the fastest computer.

[Art Dunn]

I'm having the same issue with our Pinewood Derby which goes off on March 18th. We originally had 25 scouts and a three lane track - a Perfect-N with 100 heats. Recently we had a last minute addition so now we have to proceed with 26 scouts. Splitting them into smaller groups is really not an option so I'm looking for any way to create a PPN with 26 racers on 3 lanes. Any ideas??

Thanks.

[Stan Pope]

My inclination would be to run 2 rounds on a PPN chart ... that would be 6 heats per racer. Since opponent balance it not perfect, I'd then run a Perfect-N finals involving approximately twice as many racers as you have trophies to award or places that must be determined. For instance, if you need to send 3 or 4 reps to district races, run a PN-7 finals.

The reasoning behind this is detailed on http://members.aol.com/standcmr/pwmevalc.html

[Cory]

I'm having the same issue with our Pinewood Derby which goes off on March 18th. We originally had 25 scouts and a three lane track - a Perfect-N with 100 heats. Recently we had a last minute addition so now we have to proceed with 26 scouts. Splitting them into smaller groups is really not an option so I'm looking for any way to create a PPN with 26 racers on 3 lanes. Any ideas??

Thanks.

Art, I'm curious as to where you got your 100 heat PN chart. While it doesn't surprise me that such a chart exists, Stan and I are (to my knowledge) the most common source of PN charts, and I don't believe we have that one.

On another note, for larger numbers of cars we've found that one set of generators will often work for multiple numbers of cars. If you know how the charts work, that might give you the key for the 104 heat chart you seek. If not, then I think Stan or I would be glad to take a look at your 100 heat chart if you're willing to share it.

C.

[Art Dunn]

Cory - Actually, the 25-3 chart came from Stan's web site http://hometown.aol.com/standcmr/genframe.html?f=fs. I'm very interested in this 104 heat chart you're referring to as I can only get the PPN charts to do 2 rounds on a 3 lane/26 racer setup. Is there someway to flip the cars around and basically redo the 2 round PPN chart?

[Stan Pope]

Ahh, yes! The old generator program that only worked PN charts! PPN generator much more versatile, with PN gaps filled by PPN charts (except for a few very difficult cases.)

"Flipping" PPN charts produces inferior charts ... the "1" max variance in opponent matches becomes "2" almost all the time, dropping them into the category of "Miscellaneous" which fill the few gaps in the spectrum of PN+PPN charts.

[Art Dunn]

Yes, that's true. I've been trying to somehow randomize the cars and re-map them to the lane assignments and I'm having a heck of time doing that. Is there anyway the code for generation of the charts be extended to do something like a 4 round chart for 26 boys on 3 lanes? If the code is available to be modified or more CPU is needed (I have plenty), maybe that might work?

[Stan Pope]

Yes, that's true. I've been trying to somehow randomize the cars and re-map them to the lane assignments and I'm having a heck of time doing that. Is there anyway the code for generation of the charts be extended to do something like a 4 round chart for 26 boys on 3 lanes? If the code is available to be modified or more CPU is needed (I have plenty), maybe that might work?

The javascript is there, but it is a tough change.

Cory has a new version of the DLL implementation that maps "racer number" based on a randomized table. Generate two PPN charts with randomized mapping tables, and combine them to get a Misc. chart that is twice as big. Hmmm ... Cory, your spreadsheet might be just what he is looking for ... modest changes in the spreadsheet would produce the 4-round chart that he is looking for. Only works in Excel but lots of folks have that. Hey! Is that version online on the Pack 146 website, yet?

[Cory]

Cory has a new version of the DLL implementation that maps "racer number" based on a randomized table. Generate two PPN charts with randomized mapping tables, and combine them to get a Misc. chart that is twice as big. Hmmm ... Cory, your spreadsheet might be just what he is looking for ... modest changes in the spreadsheet would produce the 4-round chart that he is looking for. Only works in Excel but lots of folks have that. Hey! Is that version online on the Pack 146 website, yet?

The spreadsheet and DLL are here:

http://pack146.nova.org/pinewood/software.html

The spreadsheet code is "protected" but I have no issue with giving out the password to anyone who is interested in making improvements.

The changes aren't too bad. Let me think on it, though -- I'm wondering if there might be an even better path to take.

[Stan Pope]

The spreadsheet and DLL are here:

http://pack146.nova.org/pinewood/software.html

The spreadsheet code is "protected" but I have no issue with giving out the password to anyone who is interested in making improvements.

The changes aren't too bad. Let me think on it, though -- I'm wondering if there might be an even better path to take.

Oh, oh! Cory's getting another "brainstorm." Man the lifeboats!! Women and children first!!!

Seriously, doesn't surprise me.

Hmmm ... Is he adding another parameter to the DLL interface??? an Iterate factor?

[Art Dunn]

Cory - I see the anaysis tools but I don't see the spreadsheet anywhere. Can you point me to it?

[Cory]

Cory - I see the anaysis tools but I don't see the spreadsheet anywhere. Can you point me to it?

Bottom of the page...here's the URL for the ZIP file in case it's not appearing for you for some reason.

http://pack146.nova.org/pinewood/ppnxls.zip

There are also a couple of screenshots.

C.