In a message dated 7/7/00 9:52:00 AM Eastern Daylight Time, Ben Thatcher
writes:
> Good point GH. The simple part isn't the startup, it's the ongoing part
> of it. I think looking at some results would be a good place to start and
> work from there.
Be my guest. I'd REALLY like to see where you'd start. Making older cars
more competitive is one of the goals of this sort of classing method, right?
How do you decide where to put cars that were previously popular in Solo II
but haven't been seen at a National event in the hands of a good driver for
several years? This is the reason the SEB is so hesitant to drop such a car
into a lower class. What's your frame of reference when the top drivers that
were previously in those cars have moved on to newer ones, and you've never
seen the older cars run against the latest ones with comparable drivers? How
do you do that by "looking at some results?" The ones you need to look at
don't exist.
> Then after the process is started you can see which cars aren't
> being entered in the classes. More than likely, there's a good reason.
> Drop 'em down for a year and see what happens. Keep dropping them
> down until people start entering them and winning. The same thing would
> work the other way, a car dominates a class, move it up a class and see
> what happens.
What will happen as soon as you start dropping cars down a class or two
is the creation of a Car of the Year sweepstakes in every single Stock class.
There are a number of people who are pretty good at assessing the autocross
potential of a car, even one that's never been seriously autocrossed before,
and these people will take one of those "dropped" cars and make it a class
overdog once the opportunity presents itself. When that happens, what do
you do? Move it back up immediately? If that's the case, very few of those
cars you weren't seeing will ever appear in the first place, negating one of
the primary goals of your system. No one will invest the time and effort to
develop the car if their advantage can be taken away so easily. What if you
leave such a car in the class as a clear overdog for an entire season, then
re-class it? No problem, that just means that the folks who believe you have
to have a brand-new (or in this case, "brand-new used") car each and every
season in order to win will finally be right. And how would such a system be
beneficial to regional competitors?
> Let's face it the SEB is far from 100% in classifying cars now and this
> process is already in place to some degree. However, trying to keep
> similar type cars together in a class has its problems, too.
And your system, which would take several years just to correct for your
initial classing mistakes across the entire Stock category, would be more
accurate? All those class results from past events under the current system
are out the window as soon as you start moving cars around based in part
on which ones do or don't appear at events, so you're building your database
for making performance-based adjustments from scratch. And how many
cars get fully developed for National-level competition in just one season?
You have a limited number of events during a year with which to compare
car/driver combinations, and you're going to shuffle all the classes after
every season? You can't assume that the top drivers will stay in the same
car long enough for you to accurately evaluate it, because they have no
incentive to do so. The incentive to find a "ringer" car every year will be
overwhelming.
> One of the main problems with the current system is that HS is similar to
> ES and DS. FS and AS are similar, BS & CS are much alike, also. So, when
> you have to move a car now, you have to move among those similar type
> cars.
>
> However, the actual times between classes are all over the place, not
> linear from SS to HS. Classing by time instead of car type would help
> some popular cars that just don't quite fit in, now to be competitive.
Assuming this is one of the methods you'll use for moving cars up or down
in the class structure, what sort of time margin do you consider
to be "all over the place"? One second for two days at Nationals? A half-
second? A half-second over two days would put AS and BS together and
CS and FS together. You're not likely to get a mix of all the cars in each of
the paired classes, because most of the top drivers will pick the car with the
half-second advantage. You won't get a more varied mix at Nats, you'll only
get a different "car-to-have for the class". And again, how does this benefit
anyone at the regional level? A really good driver at the regional level can
win in nearly any car he drives, so the perception of some that their own car
is misclassed will continue.
You still have only 9 Stock classes in which to place hundreds of cars, and
even if your performance-based system results in classes with more closely-
matched cars time-wise, most of those classes will still have one or two cars
which will show up at Topeka in the hands of the best drivers. If you class
some of those cars that "don't quite fit in" competitive-enough nationally,
they
will simply displace one of the current "cars-to-have" in whatever class they
may be in. I still don't see how the end result is any different from what we
have now. For someone like yourself sitting in Street Prepared it would be
really interesting to see all the different cars come and go, but I don't
understand how it would benefit Stock category competitors at the national
the regional level.
GH
|