Since some posters seem to be telling me what my position is, even after I tried very hard to not have one here...
I'm frustrated by the strategies being applied to argue what I think is a completely reasonable solution to our "process problem," by invoking counters to positions that I am NOT proposing (e.g., blind adherence to one simplistic formula for all cars, the addition of an endless list of factors to a psycho-complex process, or entertaining repeated requests for minute changes to specification weights).
So here - as simply as my addled mind can manage - is what I'm arguing for:
** Start with the current process, exactly as the ITAC currently applies it (we have written but not completely fleshed out guidelines).
** Document any and all power multipliers that are in play, by describing what physical attributes warrant them (e.g., '70's era cars with smog pumps and carbs get 1.xx; OBDI/OBDII FI 8 valve cars get 1.xx, etc.) If the ITAC thinks more are necessary, add them. This list would likely be 1.25 unless (whatever)...
** Document guidelines for any and all adders that are in play, also by the physical attributes that trigger their application. FWD is binary, others will be harder. If gearboxes are on the table, provide example ratios of what makes a 'box "good." How big are brakes that deserve an adder? Document it. NOTE here that I am NOT advocating or the inclusion of any more factors.
** If it ain't in on the lists above, it doesn't get included in the figuring. If we discover some new thing (hey, how about ABS?) then adders/subtractors or new multipliers (the Prius, you know) get listed and used. There's no ad hoc addition of fudge factors or new variables, tied to individual make/model examples.
** When a new car comes into the system, its pertinent attributes are documented (there's talk of creating a new IT-specific VTS) and the math is completed. STOP. Do not second guess, do not re-run the math with different factors to see if a more comfortable/palatable solution pops out. Do not buy what individuals with vested interests are selling as "real world data" on individual cases.
** Document the assumptions and math for posterity (including the date), publish the math, go racing. The documented examples will help inform the standards used to apply tougher adders, since we'll have something to compare to.
** When a request to re-examine an existing car comes in, run exactly the same process. The only difference is that it MIGHT be reasonable to have a tolerance, within which existing weights are left alone if they are "close enough," because there are costs ($$ and otherwise) associated with any change. This (of course) has to be recorded so it gets used consistently.
** THe ITAC might trigger the process ourselves, if we identify an anomalous listing. I don't believe at this point that it's truly necessary to do a Second Great Realignment. Arguments that it's "all or nothing" honestly strike me as somewhat hyperbolic.
That's it. Done.
** We won't have an ongoing flood of requests to fix cars that have been through the process, because as soon as members understand that we've documented the process and outcomes so they get the same answer over and over, they'll give up right quick.
** We aren't trying to be "too smart." In fact this system's got a lot less hubris built into it than does the most subjective applications of the process that have happened over the years. I don't believe we can subjectively get closer to "right" than this process will achieve because "what we know" is too suspect.
** We can - without concern - publish this process to the membership. Since we aren't making anything up as we go, we have repeatable. Since weights get assigned to attributes rather than cars, we have defensible. Since we can communicate processes to the membership, we have transparent.
** The ITAC would lose some degree of control to apply their judgment. I don't care. The value added in terms of trust and reduced conflict is worth it.
** Yes - we are still going to have some people say, "Smogged up POS Pintos, Vegas, and their ilk should be at 1.xx +.10. "Thank you for your input." Yes - we will still have some issues where categories include ONE car (see also, "Rotaries"). The ITAC will just have to explain - a few times, perhaps - that it's gotten as close as it believes reasonable. If TRULY compelling evidence becomes available, the ITAC has the power to change a factor but we'd be well advised to have some standards for evidential quality for that step. We also better be DAMNED sure about the change because it will throw past classifications into inconsistencies with those done post-change. This kind of thing should be VERY rare. If they ever happen, they should be documented as changes and dated.
We members of the ITAC would have to hang onto the conviction that we're doing the right thing and have the stones to explain it and stand by it publicly.
Argue what I've proposed but please stop telling me that it's stupid to do something that I've never suggested.
K
, but if you want to complain about the numbers you need to talk to Mazda.
