AspireTec Engineering  468
PrevPrev Go to previous topic
NextNext Go to next topic
Last Post 17 Apr 2013 06:37 PM by  jrubins
M3 swap e36 325is class
 44 Replies
Sort:
You are not authorized to post a reply.
Page 3 of 3 << < 123
Author Messages
mrazny
Basic Member
Basic Member
Posts:462


--
15 Mar 2013 11:17 AM
The problem with the data out there is it's time only... You wouldn't know what's wrong with certain cars, or which cars didn't do a full tilt build, and the biggest one, the driver talent. The data has little control.

Besides having a good system of point assignment, and how to evaluate and change that point assignment, the workload the system would need is crazy huge. Just look at the index for Stock Classing. You'd have to figure out the base points for all of those cars, figure out which cars are too slow for the class so that they have a smaller starting point, etc.

Then you'd have to figure out how to manage the mechanic of engine mods.

A car has 170whp stock. individually, all these parts are dyno evaluated to give; Cat-back exhaust (+5hp), Headers (+9hp), pulleys (+6hp), ECU chip (+13hp), intake (+4hp), Race Cat (+2hp). But lo and behold all together it doesn't net the sum of the parts does not get +39hp to get the car to 209 hp. Nope, the sum of the parts gets it to 196hp.

Money and testing will always be a part of some racers process. So when you put a points value on certain mods, let's just say a point per HP alone, but the next class is less than 39 points away, some guys will have the time, resources and money to optimize those mods to stay under the threshold and others won't. Some guy will find the perfect combination to spend only 23 points to get actual combined 20whp and stay in the lower class, some other guy will spend 26 points, only get 14 whp, get bumped to the next class, then rant and rave about having less power than the guy who didn't get bumped, so why did he get bumped.

More to the point, running a region has so many variables already, including lamenting over the state of the rules, the classing, Street Tires, the .975 modifier... I think the points system would have a negative effect on the regions ability to manage a day-of, because there are 7 broad categories to present and that's it. Each has their allowances. The points system is going to be a specific to your car, and a specific to every one of your mods you have, and a specific to every mod you haven't done to your car. The specific will trigger more of a response from each competitor. Some will have strong logical positions of conflict. Some will just believe they are being wronged with no actual data or logic to back it up (a 19 yr old kid couldn't possibly beat me, etc.). You won't have controls.

After a certain point, you'll have to put a value on a suspension type vs. a different suspension type (which is dynamic).

The same mods to an e36 M3 through the present levels of classifications we have won't net the same reduced times as those mods to an RX-8.

From my standpoint, it's too many variables to knock out when we are trying to stay a low-cost, introduction racing. You could try some sort of power/weight thing, but then you'd effectively require dyno time for every competitor, and scales at every event.

BMW doesn't do a national point system, each of their regions kinda do their own thing for autocross. The points system I've seen are still broken, even when you have the control of at least similar cars (mcstrut front, multi-link rears). And being a regional where the decision makers are out in the open, it has even more inertia than the SCCA model out of fear of change. The 135i in stock form is basically in their highest point category, with fully built M3 Road Race cars. That's with an e-diff, body roll to infinity, and skinny front wheels. They tried to "point" out the difference between street tires and race tires etc., it doesn't quite pass the smell test. It can still be fun because its autocross, but it didn't bring more parity even within a single marque.

The points system could work with a TON of modeling, resources, money, time. I don't think from the no resource, just a few guys "thinking" and looking at times that the system could work. What would you use for the difference between 140 UTQG and Hoosiers? A .975 modifier to the points? I think that one is too dynamic to quantify.
mrazny
Basic Member
Basic Member
Posts:462


--
15 Mar 2013 11:34 AM
This isn't to say don't think about things, but this is to say don't assume that the system that you think is broken is being run by 30 guys who don't give a sh*t. There's a lot of nuance, and a lot of people with quality thoughts coming from the right place, trying to improve the sport. It is that much of a challenge to just keep things going as well as they already are.

I don't mean "just take the hand you're given", but it didn't come across like you respected the challenges and totality of the club as a whole. Yes some of it definitively IS inertia. And that comes from a simple dichotomy:

We have people that come out and support the SCCA. How many of those people would be chased away by this change? Like the Hoosiers vs. 140 UTQG in Stock debate. Of those 300 or so Nationals competitors that come to race Stock on Hoosiers, if you change Stock to 140, did you lose them forever or just make them annoyed but they'll still race.

On one hand that's 300 of 1200 at one event, and you most likely would not lose all 300, and may replace some of them with supposedly happy 140 UTQG people, but it's a big thing to measure. The other worry HAS to be, how many of those that throw up their arms and walk away from SCCA Nationals would walk away from SCCA Regional? And how many of those people are board members or other specialized workers that keep the events running? I know from our region there'd be 4-5 very crucial people that would consider walking away. The hope is someone steps up, but that's not a known quantity. Yes that reads as the old guard holding the sport hostage, but it is a reality when we're volunteer.

The leadership isn't dismissive, it's just a huge job to juggle. A provisional side class to try out a points system isn't the worst thing in the world, but you wouldn't have the data you'd want with 10-12 people interested. And if you hold a 100 person region hostage to an experiment, you might shrink a bit from that 100. It might not seem that scary, but if our region dropped to 70 people at each event, we'd be hemmoraging money, probably have to raise entry fees, then lose more people to other local clubs since we're already the most expensive club in town.
btwyx
New Member
New Member
Posts:69


--
15 Mar 2013 05:28 PM
Posted By mrazny on 15 Mar 2013 11:17 AM
BMW doesn't do a national point system, each of their regions kinda do their own thing for autocross. The points system I've seen are still broken,

When I looked at the BMW points system for our MINIs, I was surprised to find that the difference between our HS MINI (Cooper with 118hp) and our CS MINI (JCW with 208hp) was 3 points. By their scoring system they expect 1 point to be about 0.01s (0.03s in total) over a minute run. PAX would give the HS car 1.6s.

I'm not sure I believe either index system, but PAX seems closer to reality. If I were to run a BMW club event, I'd be taking the JCW, rather than the HS MINI we use for SCCA.

OZMDD
Basic Member
Basic Member
Posts:131


--
17 Apr 2013 03:30 PM

I have to say that the OP has not done himself or his cause any favors by the tone and content of his posts. From my perspective, you come across as woefully naive about the sport, which makes you a poor advocate for a ground-up classing system. It would be like me suggesting a new scoring system for olympic gymnastics (bad idea).

I don't actually know you, so I won't claim to know how you really are. Your two arguments are absolutely horrible (M3 motor swap and C5 w/ shifter) because they illustrate the wrong end of the spectrum. A swapped-motor BMW does not represent even a tiny fraction of people who come out to try autocross. Most new participants own a relatively-new sporty car that they want to get more enjoyment out of. Trying to estimate the performance potential of a motor swap is beyond difficult. I would happily put a 94 Miata w/ ABS and torsen and a swapped-in 2012 Miata drivetrain up against a C6Z on Hoosiers in a typical regional autocross course and wager heavily on the Miata. Can you do the math to determine the points for that? I can't.

The example of a C5 with a short shifter is a red herring. No local autocross has ever dumped a noob into an impossible class based on something that benign. They are generally welcomed, told to run in Novice, advised that their minor mod is not class-legal so they might want to change it or prepare to run in a tougher class in the future, and told to have fun and not be too upset when the dude in the Yaris spanks him by 10 seconds.

It is common practice at local events to inform your competitors about non-legal mods and ask their permission to run in their class anyway. If they are deemed to provide no meaningful competitive advantage, you are good. If you kick everyones butt, you will be politely asked to reclass at your next event. Unless you are an alien or in a very weak class, you won't have to worry about that for a while.

Again, the tone of the OP came across as "these rules are too complicated for me and my car doesn't plug-in easily, so change everything." If that's not accurate, then start a new thread with a clearer message. I do have to disagree with your assertion that it ought to be easy to show up and have your car be competitive in a class. There are far too many cars out there to EVER create a system that has a place for all of them. This is motorsports, not kindergarten, and we don't do participation ribbons.

Welcome to this!

jrubins
New Member
New Member
Posts:19


--
17 Apr 2013 06:37 PM
Hmm, it seems I missed some posts. I wasn't notified by the forum software until OZMDD's post today. Weird.
OZ, it's interesting that to you I seem 'naïve', but others thought I was being sinister by hiding my background. As I pointed out earlier, I have a few years of autocross experience.
No doubt the M3 swap is a bit of a corner case, but if you think about hondaworld - there are many swaps to build a higher spec model. I wonder how many "god's chariots" are really DXes with the Si motor dropped in. Maybe none, but it's worth asking since there are tons of swapped Hondas in the world.

mrazny, it's not an easy task to figure out exactly what mods are worth what. It's doable though. Given enough data, I could figure it out.

What little time I have, I did do a quick ANOVA to see what classes are signficantly different (statistically speaking) using Nationals data. 2 courses, 3 runs each gives a fair bit of data. It turns out there are quite a few classes that aren't statistically significantly different.

Out of 3916 class-to-class comparisons, only 1135 were different at the p<.05 level, 1222 at the p<.10 level, 1372 at p<.25, and 1532 at the p<.50 level.
What that means is that 2384 class-class comparisons are not able to demonstrate they are different with 50 percent confidence.

The ANOVA statistic isn't really the right test to run here (for you stats nerds, it should be a Chi Squared), but it does indicate there's hay to be made. With so few classes being significantly different, we can be reasonably certain that some (if not many) classes are overlapped.

What that means is that classes are doing a poor job separating competitors into competitive groupings. What we'd expect with a fair, useful classing system is that competitors in one class will be grouped reasonably closely together, and competitors in a different class should be further away.
That isn't the case now.

There's still plenty more research to do, but I'll keep hacking away at it as I get time.

Cheers,
jrubins


You are not authorized to post a reply.
Page 3 of 3 << < 123


Woodhouse Motorsports SPS 88x31 Button
G-Loc Button Vorshlag 88x31 Button
Leroy Engineering Micro Button Sunoco 88x31 Button

Advertise on SCCAForums.com and reach thousands of visitors per day!

SafeRacer FREE SHIPPING over $99

Shop for Pirelli tires at Tire Rack. blank



Sunoco Bottom 468x60 Banner