Do I recall that marketing campaigns will play a part in the final game?
If so, I was watching one of Killrob’s videos where he is explaining the n-dimensional Pythagoras method of calculating the engineering time.
It got me to thinking - a marketing boost could work the same way. Because a marketing campaign is about highlighting all the strengths and not the weaknesses, if the game were to compare the stats between your car and the best competitor in that stat it might work in the following way.
Any category in which your competitor outperforms you, you leave off the ad; so the negative measurement is limited to zero.
However, let’s say the only two things your car is better at than a competitor is 0-60 time and fuel economy and the competitor trounces you in every other category, you could still plaster those figures over your ad campaign and win over a few buyers who care about those things.
If it was only one thing your car does better, let’s say just the 0-60, at least you have something to base the campaign on and leave a snappy impression.
Using Pythagoras would give diminishing returns that wouldn’t explode the number absolutely if your car is better all-round, but also it would reward one thing being a LOT better and other things being average or even worse than the competitor - just like a good old sneaky ad campaign in real life.
But specialising too hard in one attribute would leave you comparatively open to another competing car taking your 0-60 crown and stealing your hard-earned buyers.
In fact if you imagine one car which is an all-rounder, the ad campaign will be more resilient because even when a competitor exceeds one of those categories, it makes very little difference to the overall advertising boost score. What you would be terrified of is another very slightly better or otherwise closely matched all-rounder coming along and eating into your orthogonal magnitudes. That would be the way to do it!
The upshot would be that the marketing boost system would reward cars with more extrene but variable stats for shorter production runs, but would reward all-rounders for cars that will have a longer life cycle. This could be an interesting dynamic I think.
Come to think of it, something like this with a degree of randomisation might be a good way to model which cars certain customers buy, but that would certainly be a different subject for a different day!