How would you figure out how many 1975 roadsters with L-82, 4 speed, A/C, 2 tops, power windows, tilt/telescopic, dual horns and FE7 were made??? Is there a way to break it down or close estimate???
Thanks

 

You can't. GM did not keep data on options configurations. About the only thing you can confirm is that 4629 convertibles were assembled, 2407 of those had the optional hardtop. The car you mention would be one of 2407 provided the hardtop is original.

 

I've developed a far more accurate method of calculating the answer for these how-many-did-they-build questions. We all know that for bragging rights, we want the answer to be some very low number like '1 of ten' or less. If we use a bit of logic we can also figure out that virtually all Corvettes are actually '1 of 1' due to the low volume built and the wide variety of option available even in the most austere years. We also know that we can't all claim the 1 of 1 status at the same time because nobody will believe us. We already have a nasty reputation for being one of those '________' Corvette people (insert favourite adjective) so all of us claiming some outrageous rarity number will just prove them right.

Here's how it works- think of the day of the month that the two most important people in your life were born. Include yourself if you like as one of them. Almost invariably, the two people will not be born on the same day. Take the two numbers and subtract the smaller from the larger. That number is the '1 of' that you've been seeking.

Ex- one person was born on the 22rd, the other on the 18th. 22-18=4. Your car is 1 of 4.

For the rare case that both people were born on the same day, replace one of them with the number 15 and use the same formula.

If you really don't have two favourite people in your life and don't want to lie, there's an alternate method. Think of how many people have come up to you and told you the same old story like 'my uncle had one with a metal body', 'you should have got an LT-1 big block' 'I'll give you $5K for it right now' etc.

If the number exceeds what you want to hear, divide the total of one dumb comment by the total of a different dumb comment.

See, simple.

 

Out of 2372 L-82 cars built in 1975, 366 were convertibles, 2006 coupes. When I was doing my '75 L-82 Survey from about 1998 to present(though not near as much), I have about 84 cars in it and saw almost a 50/50 split on automatics vs 4spds. Seems as though most cars did have the A/C & power window options too. One way to get a close theoretical estimate would be to calculate how much each option was as a total % of the total production of 1975, then take that % of 366, the known number built of '75 L-82 convertibles.

 

Paul,

Where does that 366 L82 convertibles come from?

 

Any speculative estimation on the distribution of a % of that (and any) combination of options not completely based on actual production data is completely statistically invalid.

Its pretty close to random. You cannot project numbers that result from different people making individual choices. Numbers people quote are purely made up.

Calculating the amount of cars that year with a combination of NINE (9) options/choices is imaginary. If you are trying to be able to say it's rare, just say it's a high option car that was likely relatively rare.

 

Statistically.......none

 

You calculate what percent of L82 convertibles were of the known L82 option total

 

That would not give you a known number, that would give you a statistical guess.

 

Absolutely-

Another common flaw in calculating numbers is not taking into account options that a given car does NOT have.

Ex- power brakes and power steering are very common. In just about every year, more than 50% of cars have one or both. This means that cars that do NOT have them are 'more rare'.

My car has deluxe interior, PS, PB, rear defrost but manual windows and no AC. Extra- extra rare I bet.

 

There were 4,629 convertibles produced in 1975. Just having one from those remaining to this day is "rare". Don't sweat the small stuff. Those RPO stats that you cite do not exist. Global automobile production in 2014 was ~18M. Does that put the 1975 convertible in perspective?

 

Hey Mike,

From a statistics perspective, if you multiply in a percentage for every option, whether high or low, you are accounting for it. So if 97% of cars had power steering, you are accounting for the rarity of the 3% of cars that do not.

 

But that's not how people typically calculate the 'rarity' of their cars. They run down the list of what it does have, not what it doesn't have.

 

Somehow this does not matter to me....

 

I see what you are saying. When you perform this worthless calculation, the entire options sheet does need to be considered, but people only look at the actual options they have.

 

Exactly, but I don't think the calculation is completely worthless. Maybe everybody should do it and discover like I did that long before you get to the bottom of what it does have, never mind what it doesn't have, the calc will show that the car is statistically less than 1 of 1.

 

Well written! How many cars are there in the USA today? And what percent of those are of the 4629 75 convertible . Not a drop in the bucket in comparison.

How many of those 1975 convertibles even exist anymore ......probably less than half anyway if that.

I never saw but a scant few new back in the day.

 

I don't think it is that simple. I think alot of people approach this problem simply by multiplying all the percentages, basically they are using a A∩B∩C∩D.... I think the true statistical way to approach it is to calculate the probability of the combination, and you must consider that repetition in combinations is allowed.

 

Yes, correct. It is most probably true that people who ordered power steering also ordered power brakes (and vice versa) and that this group was probably more likely to also order deluxe interior- and on and on.

All of this could improve the accuracy of an estimation by indicating 'more rare' or 'less rare' but if the crude and simplistic method I outlined above already indicated a less than 1 of 1 probability on a consistent basis, is there much point?

 

I think my point is that it is statistically impossible to have less than 1, so that should make you consider that 1) it is not the appropriate method for what people want, and 2) what it is really telling you is the probability of you selecting each of those options out of a hat and you would get your car. Probabilities are always going to be less than or equal to 1.

So what needs to be done is to calculate the combination with repetition. I don't just mean that someone who picks power steering may also pick power brakes. I mean take all the options available for that year and calculate the different number of combinations. Then using the known quantities of each RPO (and standard items) calculate the probability of achieving your specific combination of options (or standard items).