Big Picture of IT - Share Your Opinions

[shwah]

I agree it's probably not accurate, assumptions rarely are 100% of the time.

I just see anything other than going with a "roughly fair" default until we know better from actual dyno numbers (which we all do understand are to be closely scrutinized) as being "wild guesses" based on on track performance that aren't repeatable and consistent.

At least with the 25% default, until there is actual data, there is repeatability and consistentcy, which for IT is I think a paramount principle.

That is where we differ in opinion. It is not a wild guess to look at a car that has been competitive and conclude that it can compete (given many other variable conditions) in it's current state. However since I recognize the difficulty in defining or agreeing what a percieved competitive car is, and all of the potential negative side effects of trying to do that I suggested ignoring that fact completely.

What I proposed is essentially limit how dramatic of a weight change that we allow in a single step - regardless of percieved competitiveness. Make a smaller move, I suggested moving it 1/2 the distance, and don't make another move without obtaining more data to confirm that it needs to move (and conversely make darn sure you do make a 2nd move should any appropriate, acceptable data be presented to support it).

The idea is to act as a throttle on the rate of weight change, without preventing it from ending up at the initial process number - if appropriate. I think taking too large a step in weight change presents a risk to 'upsetting the apple cart' of a given IT class, which we all seem to agree are, from a macro view, giving us good competitive multi marque racing. I also think someone that has a car that was grossly mis classed in years gone by should not have to wait for data to prove the negative that their car can not make enough power to justify the old weight. It's a compromise of sorts.

[Andy Bettencourt]

If process says a car weight needs to change by more than x%, publish that fact and move the weight by x/2% and monitor it.

Anybody else feel this is a good idea?

[gran racing]

I get the theory that you're using on this Chris, but see several issues with it. Maybe it's workable. On the flip side, if it's truly believed that a larger weight change is warranted, then change the weight. Just because a car may have had a larger disadvantage means the should continue to? This approach would also require a change in mindset by the boards. Getting the initial change would be tough enough; getting the second half would be near impossible.

[JeffYoung]

We do differ, but there is no "right" answer here.

Two thoughts:

1. I think the quote from you below is far more dangerous than the 25% assumption. I can give you lots of examples of cars "falsely" appearing to be competitive, or not, depending on driver, track, prep level, temperature, field size, etc. It's far, far more difficult to get that right in my view than to assume 25% until data comes in showing 25% to be wrong.

2. I still think that 100 lbs or even 200 lbs -- while having an effect on performance, obviously -- isn't going to "kill" a car such that it can't run up front and be competitive against 9/10ths version of the same or similar cars. So, using the 25% assumption to me gets the car "close" - in a repeatable consistent way -- until (and if) more accurate data presents itself.

It is not a wild guess to look at a car that has been competitive and conclude that it can compete (given many other variable conditions) in it's current state.

[shwah]

But Jeff - I have taken that whole issue off the table. A few times now. Stop putting it back into the conversation.

I just think it is a good idea to have some damping on the effects of big weight changes, as long as they are still able to be completed should they be confirmed as appropriate at step 2.

[seckerich]

Anybody else feel this is a good idea?

Nope, if you have a process then use it. What you should be asking is why is it such a big move.

Was the stock published HP correct?
Is the gain over 25% ?
Do you have a clue what a 5 cylinder gain should be?
The ARRC finish is irrelevant because it was not torn down and must be assumed legal
No track record was broken and they are not dominating ITB.

#200 may be a little much percentage wise to be off on a ITB car so answer the first 3 questions and if it passes the sniff test treat it equal. Might be a little hard to get it through since it is an ITB car.

On edit, Yes I understand it was classed with the usual SWAG pre process.

[shwah]

On the flip side, if it's truly believed that a larger weight change is warranted, then change the weight.

That's the thing. We don't really know what is warranted do we? It is all based on an assumption that may be about right for the 'average' car, but there is no reason to beleive that it is in fact correct for any particular car absent data. I can accept that if the change instituted is 'reasonable', but if it is a 10% swing in car weight there is a HUGE opportunity to get it way wrong, because that is a HUGE weight change.

It may not be the right solution, but it popped in my head when I was responding to the bombshell. In other words, I'm not 'invested' in this in any way, just trying to explain the logic of why / how I thought of it.

[lateapex911]

I love reading between the lines of Steve's posts.

Going to the ARRCs or Goblins Go Steve?

[tom91ita]

We do differ, but there is no "right" answer here.

.............

2. I still think that 100 lbs or even 200 lbs -- while having an effect on performance, obviously -- isn't going to "kill" a car such that it can't run up front and be competitive against 9/10ths version of the same or similar cars. So, using the 25% assumption to me gets the car "close" - in a repeatable consistent way -- until (and if) more accurate data presents itself.

jeff, the 25% rule of thumb yields the following results for two ITB hondas

86 crx si

1.25 x 91 x 17 #'s per hp -50 (FWD) - 50 (struts and beam axle) = 1834 # vs. 2130

89 accord LXi

1.25 x 120 x 17 - 50 (FWD) = 2500 vs. 2550

since these are both Hondas and pre OBD0, etc., one would expect similar gains, right?

i do think my car is overweight per any reasonable interpretation of "the process" but apparently it is "right" per the CRB's undisclosed methods.

[shwah]

2. I still think that 100 lbs or even 200 lbs -- while having an effect on performance, obviously -- isn't going to "kill" a car such that it can't run up front and be competitive against 9/10ths version of the same or similar cars. So, using the 25% assumption to me gets the car "close" - in a repeatable consistent way -- until (and if) more accurate data presents itself.

Whoops - missed item #2.

100 lbs will make a significant difference in an ITB car, and I expect others. Remember it is not important what a 'casual' racer will see with the difference, we need to plan for the full built, well driven example. 25% only gets you close if 25% is in fact close to reality. There is no reason a driver of a +25% car will ever be inclined to share that information with you, but if you take the big step on 2 smaller steps 3 things could happen.
1. The cars were right before, and are favorably classed now. NO one sends in a dyno sheet because they like where they sit with that car, but at least they didn't loose twice that weight and end up further off.
2. The cars were off before, but do in fact make more than 25% gains, and are 'pretty close' now - the key here being that we define what 'big moves' are and we are now 1/2 of a big move at most from process. So yeah we are pretty close either way in reality. The cars are on even competitive footing and no one is incentivised to prove otherwise because they are in fact competitive.
3. The cars were off before and they don't make more, or maybe they make less than 25%. Drivers appreciate the change, and embark on development with their now more competitive car, and find that no way it makes more than 25%, and provide the data to prove it. Take the 2nd chunk off and they are probably 'pretty close'.

Using this approach, a car like Tom's that some people may be skeered of because of the imposing H on the hood, may in fact have an opportunity to be adjusted (unless there is in fact data supporting the current position) if it is off by more than 'a big move'. It could take the teeth out of some arguments against adjusting a given car.

[shwah]

jeff, the 25% rule of thumb yields the following results for two ITB hondas

86 crx si

1.25 x 91 x 17 #'s per hp -50 (FWD) - 50 (struts and beam axle) = 1834 # vs. 2130

89 accord LXi

1.25 x 120 x 17 - 50 (FWD) = 2500 vs. 2550

since these are both Hondas and pre OBD0, etc., one would expect similar gains, right?

i do think my car is overweight per any reasonable interpretation of "the process" but apparently it is "right" per the CRB's undisclosed methods.

My guess is there is a beleif it makes more than 25% gains. When I ran that math for my car and found 2131 as a 'process weight'
1.25 x 105 x 17 - 50 strut/beam - 50 fwd
I wrote a letter.

It turns out there is a beleif that my car makes 30% - which processes out to 2220. It was recommended to be moved to 2270, and turned down. Not sure what else was used in the process to give the extra 50, but ces la vie.

[seckerich]

I love reading between the lines of Steve's posts.

Going to the ARRCs or Goblins Go Steve?

Going to both. Ask anyone that knows me and they will tell you I rarely bother with between the lines. I am just being calm until I see how this plays out.

PS. Need to put 100# on the ITS RX7, that is how overweight I was when I set the track record at CMP. Forgot to pull the extra weight the wife runs in STU. I would hate to carry that extra weight on a ITB/ITC car though. Not a big deal in R/S.

By extra weight I mean the lead, not the wife. Forgot she reads these.

[JeffYoung]

Chris, enjoy the discussion, but have to narrow my responses:

1. I would suggest you are not taking on track competitiveness off the table; I think at least implicitly it is included in going to a "stepped" process v. just using the 25% assumption.

2. I'm still just more comfortable with the 25% assumption until proven wrong, and the chance that a car may be 50 or 100 or in very rare cases 200 lbs off of process weight, than a more cumbersome and time consuming stepped system. Given how long it takes us to work through even 1 car, the simpler the system the better. More gets done.

I want to hit on one of Kirk's points in regards to the Audi. At it's present weight, it would require a 40% gain in IT trim for the process to "spit out" that weight.

In all likelihood, the Audi is already somewhere between 25% and 40%, and probably less than 100 lbs off its actual process weight.

So I just don't see this car being as big an issue as it appears to be to the CRB, but they disagree and we do report to them.

[Knestis]

You do not have ultimate authority. You have delegated authority which follows a chain of command from the Chair of the Board through various levels of management until it reaches you. At any point, some level of command/management above you can say - stop.

Ultimately, the technical policies implemented by the body recommending/setting those policies must correspond with the goals and objectives of the organization.

What you seem to be suggesting is an omnipotent agency with no accountability to anyone other than themselves. Will that mean the GCR will be in Latin going forward? Will we be required to eat fish on the Friday before a race? Will Dan Brown's next novel be about the SCCA?

"Delegated authority" being the key concept - absolutely! Point being, while the BoD has the ultimate responsibility for the Club, they may not be in the best position - too insulated by levels - to make tactical decisions as detailed as how much the Whatever RS should weigh.

But that's an academic aside. In this instance the issue is less about who has the decision-making power: It's about the first principles that drive the decision, and who shares them.

K

[quadzjr]

Nope, if you have a process then use it. What you should be asking is why is it such a big move.

What if it was simply it went throug the process with an error in in the calculations addtional to the adder for "balance".

If it is an admited error I would think that the process of from the ITAC to CRB and making the change would be stream line (or atleast mroe streamline) since it is not a competition adjustment, just a math error.

Post selfishly in reference to my particular car.Still waiting for my car to drop some lbs

[shwah]

Fair enough Jeff. Edit - I still think you will only be proven wrong on the 25% from cars that are missed in one direction, but not the other. That is the flaw in the current approach.

The process I suggested is simply for any car that is speced way off of process weight, without regard to competitiveness.

To the specific example, I do think my idea benefits them in the real world. If the Audi is that far off process, but people (including me, and apparently the CRB ) feel uncomfortable taking 250 or 200 or however many pounds off, you may end up not taking any off, while with this approach you take 125 or 100 or whatever off, and evaluate from there based on data. At least then the Audi gets some consideration, and if they can prove it needs more take the second step, but they are not stuck at current weight, which is where they are right now after the recommended change was not made.

[Knestis]

... It is not a wild guess to look at a car that has been competitive and conclude that it can compete (given many other variable conditions) in it's current state... .

I'd agree with you, Chris - if (and that's a HUGE IF) we could control for all of those other variables. The problem is we can NOT. We look at "make/model" as a variable that's really easy to see: We can tell an Audi from the other stuff as it goes by on the track. We cannot however readily see a bunch of other factors that in all likelihood bear more on the outcomes that we lump together and call "competitiveness."

I'm intrigued by your "half the distance" idea, in that it serves as another "policy damper," slowing down reactionary responses. That in and of itself might have merit.

I continue to be worried however about the first assumption though: That a weight SHOULD be changed (or left unchanged) based on perceptions. Your suggestion just decreases the damage potentially done based on those perceptions.

K

[shwah]

Kirk - I'm not confident we can do that, which is why I suggest dropping the consideration completely.

In that case, what is the real risk? That a car is unusually far from 'perfect' weight. OK so if a car is more than a certain amount from process weight (which is not the same as 'perfect' weight) we take two steps to get there. Taking smaller steps makes it easier to end up closer to the 'perfect' weight IMO. I have been wrong many times in the past, this could be one of them...

EDIT - to stress one more time, that there is an important second step to this idea. We HAVE to be able, willing, empowered, whatever, to take that second step if justified and go all the way down (or up) to the assumed adder process weight if data justifies it. This is why it does not simply limit the impact of perceptions, it actually is a mechanism to allow making any change at all - to prove the perceptions wrong in a way - while stimulating data generation to justify taking the second step, or not. In the end the cars that should would loose or gain all of the weight. Some that should not may loose some weight, but that is the price to pay to gain the ability to make the others more right.

[Knestis]

... a car like Tom's that some people may be skeered of because of the imposing H on the hood, may in fact have an opportunity to be adjusted (unless there is in fact data supporting the current position) if it is off by more than 'a big move'. It could take the teeth out of some arguments against adjusting a given car.

But again, that's the point - It shouldn't be possible for an "H" to warrant a power multiplier greater than the standard assumption, in and of itself. If that's a clue that the ITAC needs more information, then they go look for more evidence on which to base their decision about what number to use. That Honda should get reviewed and have its weight set using the process already established.

K

[shwah]

But again, that's the point - It shouldn't be possible for an "H" to warrant a power multiplier greater than the standard assumption, in and of itself. If that's a clue that the ITAC needs more information, then they go look for more evidence on which to base their decision about what number to use. That Honda should get reviewed and have its weight set using the process already established.

K

OK. It should not be the case, but it IS the case. Trying to find a solution to work with what we have here, and still get where we need to go. It is a slower road, but should end up in the right place, and could prevent or mitigate the effect of mistakes made when adjusting cars with the best intentions but getting it wrong due to lack of data.