ITB - what a bunch of crap

[gran racing]

Yes, but you're the king of all racers.

[Greg Amy]

my method or calculations certainly could be wrong, as i did it in about 10min. but on the face of it i think it's pretty good.

I'm trying to digest it, but I'm still not with you. But, it was a long night...let me chew on it some more.

My ideal mathematical formula would ignore weight and focus instead on power. As Josh pointed out above, since the weights are based on power any calculation based on weight is, effectively, based on power. So, take out the middleman.

Problem is, any formula you come up with, whether based on power directly or inferentially, cannot be applied equally across all classes. The reason for this is that the end calculated performance restrictor, weight, is calculated differently for each class because of different power-to-weight goals. These goals, as I understand it, were not necessarily mathematically derived but were, in effect, chosen out of thin air (granted, with reasonable expectation of performance). Thus, any calculated FWD subtractor has to take these power-to-weight goals into account.

We do this already inferentially with different (reasonably-derived, but still out of thin air) subtractors for ITS and ITR. I'm wondering if we can't do the same thing mathematically, as a function of power and class power-to-weight goal.

Can someone share the PtW goals for each class?

i think in the end the FWD cars in the faster classes are already at lower weights because their power levels are lower, so the 50lb FWD weight break is still somewhat close to the ideal percentage based on ITA numbers.

If we were to accept that, then we need to choose the "poster child" FWD car that we believe most accurately displays the performance goals for ITA, and use that as a poster child for back-revving the formulas. Then work forwards from there.

[JLawton]

Dave, Ray,

I'm sure its been discussed previously but I don't remember the outcome. How does the Golf III fit the process (or not fit the process as you guys say)???

Having driven Kirks car, having raced an A1 and a few ITB races under my belt, I'm not seein' it??

With other claims of over dogs, I understand why they are considered that way, but with the Golf III its been more speculation??

[tnord]

I'm trying to digest it, but I'm still not with you. But, it was a long night...let me chew on it some more

basically i took the average % weight break of the popular FWD ITA cars (presumably popular in part because the weights are correct) and applied that percentage to some cars in ITS and ITR.

My ideal mathematical formula would ignore weight and focus instead on power

unless you somehow use P/W and ingore actual power output, you're stepping into a big pile of shit. i can play around with how to apply a P/W ratio, but i think this is pretty tricky given that we're using multiple different units of measurement in the same formula. i'm thinking of coming up with a factor that uses P/W in combination with the base weight before any +/- stuff is tacked on.

If we were to accept that, then we need to choose the "poster child" FWD car that we believe most accurately displays the performance goals for ITA, and use that as a poster child for back-revving the formulas. Then work forwards from there.

disagree. i think you have to use what you think is the "top performing" FWD ITA car. taking the national racing approach, you'd pick the car that performs the best at "the big race," the ARRC. so either the CRX or early Integra. i do recognize the risk of doing this is creeping performance of the class upwards, since you're starting with the top performer as the baseline, and any +/- error could produce an end product beyond what currently exists. of course if you start with the baseline, the +/- could yield an uncompetitive end product also.

[Knestis]

Valid points Kirk, but the car is light. Not over 100 light we are told, but light. I have seen the car win races against stiff competition and shatter lap records with stock bottom end - granted very well driven, but still. It's light, or other cars are heavy.

I'm on record as agreeing with you. But I will not - EVER - get sucked into "that car beats me so it must be wrong" conversations.

Two separate issues.

K

[Knestis]

...To me this is the dominant car for the class (Not a runaway that will always win, but will likely hold a lot of track records). I don't think you will find many that will argue against that. The Golf III is clearly not an oddball and is a good benchmark for where the top of the class should be maintained at a max.

Raymond

...and to continue to try to make this point, you're making an assertion based on a TINY amount of data. It doesn't hold and it would be VERY dangerous for the ITAC to make classification/specification decisions based on that approach.

AND YET AGAIN, so nobody's got any room to make suggestions of bias - I PERSONALLY agree that, based on what I understand of the specification system, that the MkIII Golf is something like 50-80 pounds too light. That's compared to the process weight (as I recall, too rushed this morning to dig out the spreadsheet), not to other cars in the class that might be at different places relative to that same index.

K

[Greg Amy]

...unless you somehow use P/W and ingore actual power output, you're stepping into a big pile of shit. ...

Do you mean the P/W goals of the class, or the actual P/W of the classified car, post-calc? If you mean the former, that's what I described above; if you mean the latter then you're getting into circular logic where functions cancel each other out. In other words, for the P/W formula (the calc'd weight pre-subtractors/adders), the W is determined as a function of the P within a specific class ratio (e.g., simplistically said, class goal is 13:1 P/W, car's power is P, ergo car's weight is 13P).

If you base the subtractor on the actual P/W of the car after calculation, you're assuming that car is: first, at the actual, detailed P/W goal, and if it isn't you then add in additional error to the subsequent calcs; and second, you're basing it on a calculated formula that already takes into account power, weight, and class P/W ratio, thus will be redundant. Why not simply remove all the fuzziness in the middle that's going to cancel out anyway, and use the two base factors you have access to, power and class P/W ratio goal?

i'm thinking of coming up with a factor that uses P/W in combination with the base weight before any +/- stuff is tacked on.

Again, since that base weight is determined by those two factors (power and class goal P/W ratio) there's no reason to worry about classified weight: it's already in there (because weight is already based on power and P/W goals). To do it with the post-calc weight adds in not only unnecessary complexity but ratio error as well.

Let's see if we can't get these goal P/W ratios and do it both ways, see what we come up with. Andy? Jake? Kirk? Jeff? Anybody?

i think you have to use what you think is the "top performing" FWD ITA car...i do recognize the risk of doing this is creeping performance of the class upwards...

Exactly on the latter point, which is why I hesitate to use the top dog. If we use the baseline dog, then we can always adjust through the range. I guess the answer to that is whether we'd prefer to risk bringing underlyers up, or top-enders down, and that's more a philosophical/category intent decision than a mathematical one.

[RSTPerformance]

K-

I am not even trying to get into the argument that everyone gives Andy about the Miata in ITA or we used to have about the BMW in ITS... All I am saying is that it is a good benchmark not to exceed with any new cars classed in ITB. IF we were to classify a car say 50 - 100lbs light as compaired to a Golf III then I think we might start the slippery slope ITA did back 10 or so years ago where you now have a ton of cars classified that could never compete at the front even though they once did win reguarly.

I thnk that the Golf III is a nonquestionable smart move for anyone looking to build a car for the front of ITB. I also think that the Audi Coupe, Porche 924, BMW 2002, Volvo 142, Honda Accord, and CRX are also strong choices if you have knowledge about the car and parts sources. I am also very interested in the Toyota MR2, Porche 914, and the Mini. Would I build a Rabbit, Schirocco, Dodge Daytona, Opal Manta or Ford Mustang? NO, not if I wanted to win.

I don't see how you can argue that the Golf III isn't at what should be the absolute max performance basis for the class... (again I am not arguing that other oddball OLD cars are not also near or at that same level).

Raymond

[Knestis]

...I will say that if a decent driver is put in other ITB cars then put in a Golf of similar prep, they'll be faster in the Golf. ...

...and finally - I'll bite: Cite the data behind your assertion.

I'll remind you that you've got three variables in the hypothesis I've quoted above - "driver decency" (we'll assume you mean talent rather than morals - hee, hee); "preparation level," and car make/model (Golf, non-Golf). EDIT - and the ex-science teacher in me has to remind you that by any fundamental application of this process, your comparisons must hold TWO of those variables constant, while examining the influence on outcomes (speed, lap times, finishing position, competitiveness, whatever) of the third.

Show your work.

K

[Knestis]

...I don't see how you can argue that the Golf III isn't at what should be the absolute max performance basis for the class... (again I am not arguing that other oddball OLD cars are not also near or at that same level).

Raymond

For the gazionth time: Your fundamental approach is flawed. You are looking at on-track performance, in your little patch, based on your limited experience, and a complete lack of real data.

(To be fair, there's very little chance that there is a way to GET the data that a reasonable application of this thinking would require, so you're not to blame.)

PLEASE trust me when I tell you that you do NOT want the ITAC to make decisions about what cars should weigh based on that kind of approach.

The "maximum performance basis" (and that's not a bad choice of language, right there) is the process that says, "We have these input variables, we do this with them, and we get this output variable: The only one that the rules allow us to manipulate - WEIGHT..."

K

[tnord]

Do you mean the P/W goals of the class, or the actual P/W of the classified car, post-calc? If you mean the former, that's what I described above; if you mean the latter then you're getting into circular logic where functions cancel each other out.

yeah i meant the former.

If you base the subtractor on the actual P/W of the car after calculation, you're assuming that car is: first, at the actual, detailed P/W goal, and if it isn't you then add in additional error to the subsequent calcs; and second, you're basing it on a calculated formula that already takes into account power, weight, and class P/W ratio, thus will be redundant.

yeah, that's part of the big pile of shit i mentioned above. not to mention you're amplifying any problems with the car not being able to achieve actual power to match process power.

Again, since that base weight is determined by those two factors (power and class goal P/W ratio) there's no reason to worry about classified weight: it's already in there (because weight is already based on power and P/W goals). To do it with the post-calc weight adds in not only unnecessary complexity but ratio error as well.

i sortof follow you on this. what i'm getting at is that i would want to come up with some *factor* that is determined by using the P/W target of the class. then if 50lbs for ITA is determined to be the ideal weight break, then you should take that *factor* multiplied by 50lbs to come up with your new weight break for ITS, and ITR. oh, and by the way, it's only fair if you ADD WEIGHT to the FWD ITB/ITC cars also, since the 50lbs given to them is too much.

i also would want to make sure to use the "base weight" of the car whenever possible before any +/- for solid axle, double-A suspension, FWD, etc is included. this is to ensure that a car that looks like shit on paper with FWD, mcpherson strut, and a solid rear axle doesn't get it's weight whittled down to being an overdog. if you determined this new *factor* using the currently classified weight, you run the risk of giving it too big of a weight break because you're essentially double-counting the other considerations.

[RSTPerformance]

For the gazionth time: Your fundamental approach is flawed. You are looking at on-track performance, in your little patch, based on your limited experience, and a complete lack of real data.

(To be fair, there's very little chance that there is a way to GET the data that a reasonable application of this thinking would require, so you're not to blame.)

PLEASE trust me when I tell you that you do NOT want the ITAC to make decisions about what cars should weigh based on that kind of approach.

The "maximum performance basis" (and that's not a bad choice of language, right there) is the process that says, "We have these input variables, we do this with them, and we get this output variable: The only one that the rules allow us to manipulate - WEIGHT..."

K

K-

I wouldn't say that I live in a little patch, or that I have limited experience, or that we don't have any real data. Sure my patch is East coast, and my experience does not include the ITAC and no data exists as it would in a "Pro Racing" world but we can be realistic...

Name one track where the Golf III has not been able to compete at the front.

and I am not suggesting that the ITAC use some other approach for classifying cars, they have done a great job including fitting in prior ITA cars as well as the Golf III. I just think the we 9the club) just needs to be careful not to shed a few pounds each year on new classifications.

Raymond

[gran racing]

But I will not - EVER - get sucked into "that car beats me so it must be wrong" conversations.

Agreed.

...and finally - I'll bite: Cite the data behind your assertion.

Isn't the same reason you believe this?

PERSONALLY agree that, based on what I understand of the specification system, that the MkIII Golf is something like 50-80 pounds too light.

I'll post more later but basically the 122 lbs of torque, brake size, and stock HP the Golf III has is primarily why I believe it's classed too light.

[tnord]

Name one track where the Golf III has not been able to compete at the front.

now wait just a second. i thought that was the point? i don't know of a single car in a single class in IT that i'd say "nope, no shot in hell that thing can be up front." and now you're using this as a negative against it?

if the situation were reversed, and there was a track that your car couldn't be competitive at, you'd be screaming to take weight off of it.

[Knestis]

...I'll post more later but basically the 122 lbs of torque, brake size, and stock HP the Golf III has is primarily why I believe it's classed too light.

The formula agrees with you, as far as that goes (and so do I...!) but you've changed the question by citing variables that we DO consider, rather than those we don't. And shouldn't.

If one driver with 20 years of experience had spent 4 years developing (say) an Opel Manta, then put a similar effort into a Mazda 626, he/she would be in a position to say, "my lap times at the Indianapolis Motor Speedway make me confident that the 626 is a faster car, at their current weights." That's controlling for two variables (driver, preparation) while looking at the influence of the third - model at a particular weight.

Kirk (who thinks it's pretty funny that he's spent so much time trying to convince the system to add weight to his own race car.

:026:

[Andy Bettencourt]

Ray,

It ain't about the car. What you are seeing locally is a top driver in a top car. Look what happened at the test day when Eric got in his dad's Volvo...it's a combination of the two. IIRC, you guys have never even been to the dyno with your Audi's. Is that still true?

At the MOST, the Golf III is 50lbs light...and that is if you don't think a beam rear axle is worse than an independent rear. Remember, the process assumes the base car is a RWD, strut-based, inderpendent suspension'd car. Adders and subtractors from there...the two most common being 'minus' for FWD and 'plus' for double wishbone.

[dazzlesa]

I would look at berans car if ITB did not run with ITS. being a 2nd or third class citizen in a race group stinks. the one thing the golf has is power. very hard to out drive or out handle horsepower.
2nd is that it is late model compared to some of the relics we run.

[lateapex911]

I would look at berans car if ITB did not run with ITS. being a 2nd or third class citizen in a race group stinks. the one thing the golf has is power. very hard to out drive or out handle horsepower.
2nd is that it is late model compared to some of the relics we run.

OK, I'm confused. You have a top notch ITA car, one that's won lots of BIG marbles, if I'm not mistaken, and an ITB car. You run the ITB car from what I've seen, yet you don't like the run group?

I'm sure there's more to the story....but I've always wondered why the A car never is seen these days.

[Knestis]

...the one thing the golf has is power. very hard to out drive or out handle horsepower. ...

If I were inclined to make comparisons, I'd say that the Accord has more based on where I watched Deuce K. drive off into the distance at the SIC. But I know that my perceptions are potentially biased, that the sample size is ridiculously small, and that even if it was "evident" that he was beating me in a straight line, that's not anything like enough information on which to base specification decisions.

Again - do you all REALLY want a system where someone like me comes back to a committee discussion, allowed to inject my perceptions (by calling them "data"), potentially influencing the weight of your cars? REALLY...? Because that's what you're doing.

K

[924Guy]

If I were inclined to make comparisons, I'd say that the Accord has more based on where I watched Deuce K. drive off into the distance at the SIC. But I know that my perceptions are potentially biased, that the sample size is ridiculously small, and that even if it was "evident" that he was beating me in a straight line, that's not anything like enough information on which to base specification decisions.

Again - do you all REALLY want a system where someone like me comes back to a committee discussion, allowed to inject my perceptions (by calling them "data"), potentially influencing the weight of your cars? REALLY...? Because that's what you're doing.

K

I can only watch the net result of power/weight, on track, but having chased Beran, Peter, and Deuce... if anything, I'd DEFINITELY say Beran has better power/weight. This despite the crappy aero/top end vs. the Hondas (which are slightly better above 90-100mph IMO).

Despite the fact that none of them can get off the corners.

Anyone else out there have any more data than me?