So, what TRULY matters...?

[Greg Amy]

- I want to TRY and get each car 'correct'

The real "root issue" here is that there is a difference in philosophy, a difference in what one defines as "correct". I'm not necessarily impugning or attacking that difference, I'm simply pointing it out.

I want a fully objective mathematical system, free of human frailties and political tinkering. While I may have faith in the current ITAC members to not let their biases and opinions interfere, I do not have that faith for the future. I am promoting an even classification in advance, ignoring the on-track results.

Andy defines "correct" as a desired end result, where the list of "competitive" vehicles is long, broad, and any one of these cars could win on any particular weekend. Andy is, by any others means, promoting an even on-track result.

I personally do not believe that Andy's goal is realistically possible without a LOT of tinkering, both in advance and arrears, and not without a massive amount of seriously transparent work. I know Andy thinks the ITAC can do it, but I - truly respectfully - disagree.

I think using a 100% repeatable formula will result in a serious overdog in every class

So do I. At which point we say, "sorry, it's up a class you go." I'd much sooner tell a 1960-pound CRX to change the "A" to an "S" long before I'd face the population saying that I and my peers are spending a lot of subjective energy trying to make everyone "correct".

so what matters here? holding to some philosophy which not everyone agrees on, or actually getting the CARS ON TRACK within the performance window?

Fine, Travis, if that's what you want, then let's call a spade a spade and put in full-up COMPETITION ADJUSTMENTS, most certainly based on on-track performance. Because that's what you're promoting!

If you want competitive parity, then let's drop the facade of 'no competition adjustments' and do it right, with such things as lead trophies, annual weight adjustments and single-inlet restrictors.

If you're goal of "correctness" is to have even on-track parity, then let's do it "correctly" instead of pretending we're not. 'Cause what you're doing now with your subjective pre-adjustments is nothing but competition adjustments with a set of frilly pink panties to look good... - GA

 

[JeffYoung]

On edit -- and Kirk, thanks for that first post. We haven't had a good IT centric discussion here in a long time. It was needed.

Given the full range of marques and factors involved it is simply not possible to have a system entirely devoid of subjectivity. As Andy and others have pointed out, the result you get is a prevalence of overdogs. There is no way, for example, to apply a IT prep horsepower number "across the board" to all cars. It's just not possible.

Moving cars up and down isn't the answer either. That in and of itself requires subjectivity. Is the 1980 lb process weight CRX "too fast" for A? Will it be unacheivably light in S? etc. etc. etc. etc.

The trick is to only inject subjectivity in the process where necessary, and as a last resort. "Where necessary" becomes a function of making sure good people get put on the ITAC. That's what we can control, and what we should focus on. We have a workable process in place that the "real world" data has shown has made a lot of chassis competitive in S and A when that was not the case before. Keep putting good people on the ITAC and hope for the best. That is all we can do.

 

[tnord]

So do I. At which point we say, "sorry, it's up a class you go." I'd much sooner tell a 1960-pound CRX to change the "A" to an "S" long before I'd face the population saying that I and my peers are spending a lot of subjective energy trying to make everyone "correct".

so then what? you move a 1960lb car up a class, re-run it through the 100% repeatable, objective, rigid process, reclass it at 1660lbs, which nobody can reach, and you massively piss off a core group because you just made them uncompetitive. OR, you move a 2750lb car up a class, which gets spit back out at 2450 through your still unflexible process, and now it's an overdog in the next class up?

Fine, Travis, if that's what you want, then let's call a spade a spade and put in full-up COMPETITION ADJUSTMENTS, most certainly based on on-track performance. Because that's what you're promoting!

If you want competitive parity, then let's drop the facade of 'no competition adjustments' and do it right, with such things as lead trophies, annual weight adjustments and single-inlet restrictors.

If you're goal of "correctness" is to have even on-track parity, then let's do it "correctly" instead of pretending we're not. 'Cause what you're doing now with your subjective pre-adjustments is nothing but competition adjustments with a set of frilly pink panties to look good... - GA

call it competition adjustments if you want, i don't care. but there's a big difference between using the most objective, best information available to make a best effort attempt at getting a car within the WINDOW and throwing around 25lbs of lead because somebody won the ARRC & IT Fest.

you're not going to get rid of subjectivity in any class and have parity. fuck, parity doesn't even exist in anything but stuff like SRF. MY goal for IT is NOT to have perfect parity, because that objective absolutely does drive you towards BS like results based adjustments. you just make the best attempt at it you can, and if a mistake is made, i sure want the ability to go back and fix a class-destroying car.

 

[jjjanos]

I want a fully objective mathematical system, free of human frailties and political tinkering.

will create an overdog

At which point we say, "sorry, it's up a class you go."

Sorry, that's contradictory. You no longer have a fully objective mathematical system the moment you move a car that is an overdog.

If the process is a FOMS, then what you have is a process that takes inputs and determines the class/weight of the car and there it sits. If the process says that a Toyota Gofast's minimum weight is 200 pounds, then reliance on a FOMS says the weight gets set at 200 pounds, period.

In the above, the FOMS clearly has problems and that would suggest recalibrating and estimating the parameters and model without the use of a Gofast specific adder. I.e. The reason WHY the Gofast doesn't fit needs to be added to the FOMS.

What you are suggesting is a FOMS unless "we" don't like/believe the outcome... and that's pretty much what we have now. (Except we only use it for overdogs. Classified underdogs just die stillborn.)

 

[Greg Amy]

Guys, I'm not going to sit here and argue this with you point-by-point. If the majority of the SCCA membership wants to delegate responsibility to a group of 6 (7? 8?) guys sitting around a virtual table, tossing around "known information" and using their own "subjective judgment" to determine the competition parity of an entire category, then have at it. I really don't want any part of it.

But I can assure you I've been around long enough to know that it never has, and never will, work long term. Been there, done that.

BTW, how's that "parity" thing working out in Spec Miata? What's that, just three different cars, right? Everyone's happy, right? I'm sure working with dozens of different cars per class, all with wildly varying engines, drivetrains, gear sets, brakes, and chassis designs is just as easy...right?

Just sayin'.

- GA

 

[lateapex911]

One comment about the whole "move it up (down) a class" concept.

Assuming the car can make weight in the new class, it will, in theory, be a pointless move. Why? Because the process assumptions that failed it in it's current class will fail it in the new class.

Think about the RX-7 in ITA. People have suggested moving it to ITB. As the owner of one, I say, "Why bother"? Here's the issue. It's too slow for A, and adding 300 pounds of weight and removing wheel width is, no surprise, going to make it slower...and lo and behold, when it all averages out, it will be in the same relative spot in B it is in A....essentially uncompetitive.

The issue is that the process fails that particular car, for whatever reason.

I'll admit, I have a problem classing a new car when there is a significant suspicion that the stock HP rating is wrong. Manufacturers do that from time to time for various reasons, (tax in Japan, Insurance in other areas, bragging rights, whatever), but the effect on the classing is significant.

An ITB car that has the HP rated low/high by 6 Hp will weigh over one hundred pounds off. If we have two cars like that, but one is high, and the other low, the difference is over 200 pounds. That's a significant number in ITB!

Now, personally, I'm mostly concerned when the number is rated low from the factory, as fixing a mistake like that is very very hard, and can result in real damage to the class, the program and the club. people get seriously upset when they get reeled back in. (Witness the huge hit ITS took, then the loss of drivers/cars/members when the E36 was 'corrected").

I see both sides of the consistent/get it right debate, and would love to have a system that could automatically accommodate for discrepancies, such as incorrect stock hp ratings. I wonder if there could be a set of standards that would need to be met in order to invoke certain non standard changes/adjustments to account for such issues, that would eliminate the human/political factor.

 

[gran racing]

So do I. At which point we say, "sorry, it's up a class you go."

How do you come to that conclusion though? It sounds like through on-track performance? If it's only though a 100% repeatable formula I can't see any other way.

I also would love to protect against the previous blow-up of IT (IMO) with the CRXs, ITA Tegs, and BWMs in ITS.

I understand where you're coming from with a long-term view. Right now I feel comfortable with the ITAC making subjective decisions, but what about the next group?

 

[tnord]

BTW, how's that "parity" thing working out in Spec Miata? What's that, just three different cars, right? Everyone's happy, right? I'm sure working with dozens of different cars per class, all with wildly varying engines, drivetrains, gear sets, brakes, and chassis designs is just as easy...right?

Just sayin'.

- GA

except the objective of the two classes is fundamentally different. in SM, the "parity" has to be exactly perfect, hence the +/- 25lb, shockhat, RP, ratio, etc bullshit year-to-year, because the target is to get all cars perfectly equal at the RO's track, while still maintaining solid parity at every track. THAT's a competition adjustment.

having a performance WINDOW as a target is different. with this philosophy, you can macro-manage a much larger group of cars, with a larger margin of error, and not get backed into true competition adjustments. there's just no way to achieve anything close to "parity" across this scope of vehicles without a certain level of subjectivity.

 

[jjjanos]

I wonder if there could be a set of standards that would need to be met in order to invoke certain non standard changes/adjustments to account for such issues, that would eliminate the human/political factor.

In a word.... no.

In a sentence... Probably not. Determining the published versus stock discrepancy will be full of human/poliltical inputs and then there would be how much more can this car get and then there would be... it has no torque versus it'll pull a train....

 

[JeffYoung]

We have a process in place that works with a fairly minimum amount of subjectivity. All we can do is put good people in place to continue to implement it. There is always a chance that the train could go off the tracks.

But that same chance exists with a purely objective entirely rigid objective process. ITS E36......ITA CRX......ITA RX7 without understanding hp potential...etc...etc...etc.

How do you come to that conclusion though? It sounds like through on-track performance? If it's only though a 100% repeatable formula I can't see any other way.

I also would love to protect against the previous blow-up of IT (IMO) with the CRXs, ITA Tegs, and BWMs in ITS.

I understand where you're coming from with a long-term view. Right now I feel comfortable with the ITAC making subjective decisions, but what about the next group?

 

[shwah]

I think the original 3 points are important, but I move consistency to the top. We should be able to look at similar cars, and their specs, and the racing weights should make sense.

I also think that we do need a mechanism to 'make it right' when a mistake is made. I don't care what the semantics are - call it competition adjustment if you like (I personally don't mind if our class, and fellow drivers are deemed important enough to justify the effort required to make the cars competitive with each other). We need to enable the club to react to the next 325 ITS situation.

What I don't want is annual specification shuffles based on the previous years results. Maybe allow a maximum of one or two 'adjustments' to any given spec line weight. Give the ITAC a chance to fix mistakes, but don't open the door for perpetual tweaking. Also control the volume of 'please run this car through the process again' requests - (speaking of, is my request dead on the vine?)

Finally - wherever this ends up - it needs to be applied to all of the cars, not just incomimg iron. IMO the Great Realignment was botched, and that is why the ITAC is dealing with so many review requests today. If the process is good enough for the cars you are classing now, it is good enough for the cars that are on the books - even if they are only 50# off (or 110# off ), especially given a method to correct mistakes.

 

[DavidM]

This is a good discussion. The issue of transparency is obviously where I've had my problems with this process. If that gets addressed then I personally would be happier.

The issue of subjectivity is a hard one and has been discussed in various guises on here before. I agree with Greg and have said similar things about PCAs being applied to cars whether they're being called PCAs or not. I also agree that just objectively following the formula is probably going to result in some cars being better than others and some cars being worse. The question is whether or not that matters. Greg says no, others say yes. If it matters then how do you go about injecting subjectivity into the process. Do you attempt to do it during the initial classification based on the perceived capabilities of the car, which are based on some form of "knowledge"? This has many issues as Greg points out and has been discussed. Do you wait and base it upon the cars performance? Which also has many issues. Do you do both?

Inject subjectivity into the process and there's always going to be room for discussion and people questioning how the car's weight was determined. The only way to prevent this is to do as Greg says and run the formula and let car weights fall where they fall. Your car can't make the assumed 25% increase in power. Too bad. Your car makes more than 25%. Good for you. There's a nice simplicity to it. Of course, you'd have to go back and redo all the cars that got subjectively adjusted during the grand re-weighting.

The purist in me says run the formula and go. The utopianist in me says subjectivity is needed to level the playing field as best we can. I personally think most people would like to see an attempt at leveling the playing field even if they know it's flawed and won't be perfect.

The key to me is to have it out in the open how a car's weight is derived. If it's thought a car can make 30% over stock because it has some whiz bang cam then put it down in writing somewhere so that people understand it. People are going to complain about a car's weight no matter what once subjectivity is introduced. If it's in the open, though, people won't mistrust the process.

David

 

[JoshS]

Yes, you should use known information in classing. That is what the multiplier is for and was used for many cars classed now and is correct.

Unfortunately, there is not a consensus about what the multiplier "is for." That's what the argument is largely all about that prompted Kirk's post.

 

[gran racing]

Your car makes more than 25%. Good for you.

What if we have two models where this happens on a more significant manner, such as a low published HP number and is quite receptive to engine builds? This stands the chance of really hurting a class.

How would the S2000 have been handled? Should a strict 25% performance gain have been used when it is well known how close to the max those engines already are stock?

I pretty much agree with what Andy said.

 

[JeffYoung]

Ditto. I see the best case being:

1. a strict process, with

2. a limited amount of subjectivity to fix obvious and proven erros; and, perhaps most importantly

3. an ITAC culture of restraint and limited use of the tools to fix those errors.

 

[JeffYoung]

Same here. I agree with Dave/Andy/others.

And what about, for instance, the subjective factor for front wheel drive? We all acknowledge that FWD is a handicap, more so as power increases. Ok.....so we just discount that entirely? Known factor. Hurts cars in S.

So we stick to the process and take out the weight break given to the FWDrivers?

 

[tnord]

if you want the IT classification to reach it's maximum potential in terms of participants and success, a subjective element must be present.

the subjective factor contributes to the core elements that make up the Improved Touring "identity."
1) diversity. lots of cars wouldn't be classed at remotely competitive weights, so people wouldn't build them.
2) relative equal performance. read: no overdogs. without subbjective factors you end up with one or two car classes, and the class looks far less intriguing to outsiders, in turn, smaller grids.
3) long term stability. heck, i have more faith that IT will be around in the same basic form in the future than will SM. this makes IT an attractive place for people who don't want to change classes/cars every couple years.

bottom line is that i think you'd have smaller grids, less diversity, and a more uncertain future if the subjective factor was removed.

 

[Bill Miller]

OK, well, I think we're back to the old argument that I've had with more than one or two of the ITAC members over the years. And my position hasn't changed: pure objectivity is key, let the chips fall where they may.

ANY time you try to "adjust" the process based on "known" information, you're actually breaking the process. Why? Because you're making decisions based on what you THINK you know. Problem is, you DON'T HAVE ALL THE FACTS. In reality, you have very few of the facts. So, what you end up doing is chasing the outliers, when - in fact - you haven't caught the real outliers, you've only caught the outliers that weren't smart enough to play the game.

Then, you want to pretend that someone is being honest when they say they've done ALL THEY CAN DO to get xxx ponies out of an engine, and want an adjustment? Yeah, right...

So you are, in effect, making competition adjustments based on results, before there's even any results. Those initial "results" may not be on-track results, but they're results nonetheless. You're trying to pretend you're not doing comp adjustments, when you really are. And then you're not backing that up in the end.

You're trying to pretend you can predict the end results without actually correcting those mistakes that will inevitable happen.

Blech, blah, ptooey!!! Honestly and frankly, guys, don't know SHIT, you only know what people choose to let you know.

Sorry, but you're not nearly as smart individually or collectively as the whole of the population you're trying to "govern". Central Planning doesn't work guys.

Plus, you won't use the intestinal fortitude to back up that Central Planning by adjusting based on on-track results...so, in effect, you're trying to govern the outcomes without actually governing the outcomes.

It. Will. Fail. Always has, always will. - GA

I think you pretty much nailed it Greg. And here's an important line from Kirk's post

Note here however that I'm specifically talking about placing cars in classes and determining what they should weigh - not about changes allowed by the rules.

Please take note of the bolded section.

Go with what they should weigh, based on published data (from the mfg) on a stock example. Trust what the process says the weight should be, and go with it. The mechanism is already in place to correct variances and bring outliers closer to the middle, it's called PCA's. This is what should have been done during 'the great realignment'. All cars would have been considered newly classified, and PCA's would have taken it from there.

As Greg said, once you start introducing subjective inputs into the initial classification model, you're making defacto competition adjustments before the car even turns a wheel. Addressing cases where the process doesn't "get it right" are already built into PCA's. Use that!

And I'll extend the consistency component a bit further. Similar cars w/ similar characteristics should be in the same class. Either that, or create dual classifications for those cars that could go either way, and let the membership decide where they want to race them.

Objectivity is key, until such time as there is sufficient evidence to demonstrate that the whole is not the sum of the parts (either up or down). An important aside to this, is that once a variance is identified, it needs to be addressed in a timely manner.

 

[JeffYoung]

Bill, good to see you posting again.

 

[shwah]

I like what Bill said.