Wow, calm down there, man.
Maybe the maths are better than what you can understand, thus the other cars scored more for that public. Maybe the added sportiness wasn’t worth the sacrifice in comfort or the price gap, it can happen in real life too.
And also, if you really thing the algorithm isn’t good, do your own, better version then. Good luck with that.
I’d also like to point out that the figures for what’s important for each group were changed from CTC 1 to CTC 2. This is clearly a work in progress to tune things (which, BTW - thank you, Der Bayer, for putting these on).
Right off the bat I see a couple things that I personally think could be adjusted. And it’s entirely possible that I’m wrong on these, but it’s just my opinion.
First off, the fact that any GT or Super Car buyer bought my car (or NormanVauxhall’s, etc.) tells me that there are either too few high end cars produced or too many supercar customers in the simulation. I noticed no Compact (an in many cases no mid-) buyers touched the Supercars. This would accurately reflect the economical prowess and preferences of people who typically buy those cars. Maybe with a little adjustment the same will be true the other way around.
Second, it can be that your own guess as to your target audience may be way off. I was aiming for the $ and $$ compact categories. Boy did I miss. My best buyers? Transport, family, and offroad $. Even got some mid-$. I’m guessing that odd combo is thanks to my suspension setup. Looking at your car’s figures, I have a hard time placing it. It’s almost like a mid-sport that’s priced to undercut its competition. But by too far. You did reasonably well in all of the performance car categories.
Should you have been outsold in the hot hatch category? I’m in agreement with you. I think not. But again, maybe just my thinking. Though Der Bayer showed us the charts of what buyers are looking for, I have no clue how his algorithm ACTUALLY works. Maybe there’s something in there that we’re missing that says that the little cut-rate cars ARE more appealing. Or maybe it’s a glitch, that the simulation swung wildly from one end of the spectrum to the other as far as consumers’ attention to price and operating costs. And if that’s the case, I have faith that he will work to correct it. Expressing your concerns and reasoning is constructive. But losing your mind over it isn’t.
It’s all part of the adjustment. I have no doubt that the next challenge will have more tweaks to it. After a few more I’m sure the lines between categories will be much less blurred. Cars like yours will have little to no competition from the cut-rate econoboxes that are currently dominating production.
Boy, my car didn’t appeal to ANYbody! Back to the drawing board. I did, at least, touch the market(s) I aimed for. I just didn’t do very well in them. 
Welp, I guess a car company can’t survive on supercar sales alone. Didn’t do badly in that segment, though I had hoped to pick up some of the GT and trackday sales as well. Shouldn’t have tried to broaden the appeal of the car, stripped out interior and strung the engine out to the limits of reliability maybe could have hit high 50s sportiness and got some more sales. Either that or not tried to sell it for 480k, could be that too I suppose…
Thank you again for running this Der Bayer, eagerly looking forward to the next one!
[quote=“VicVictory”]
It’s all part of the adjustment. I have no doubt that the next challenge will have more tweaks to it. After a few more I’m sure the lines between categories will be much less blurred. Cars like yours will have little to no competition from the cut-rate econoboxes that are currently dominating production.[/quote]
My car was not exactly cut-rate. I am not quite sure why it did so well in the compact $ and $$ since I aimed for Mid $ and $$ and as a result made it nicer and more expensive.
Looking at it I think I know why my car did surprisingly well in the hot hatch section and it has everything to do with my focus on the mid range. I did entirely ignore sport and prestige focusing on 4 things, tame, comfort, safety, and utility.
So sport has a weight of 7. Tame + Comfort + Safety + Utility + price have a weight of 21. The other categories added up outweigh the sport rating by 14 points. So by focusing on sport at the expense of everything else cuts you off at the knees as the customers still want more then just sporty.
As to exactly how I decided to design my car that is a company secret. But focusing on only 1 stat at the expense of the other 8 does not work.
[quote=“Leonardo9613”]Wow, calm down there, man.
Maybe the maths are better than what you can understand, thus the other cars scored more for that public. Maybe the added sportiness wasn’t worth the sacrifice in comfort or the price gap, it can happen in real life too.
And also, if you really thing the algorithm isn’t good, do your own, better version then. Good luck with that.[/quote]
[quote=“NormanVauxhall”]Well, price have the same wieght of sportiness… and don’t forget the running costs too. Both have a important weight on the customer behaviour. We are still in the “cheapy” hatch market.
Mine car was near a third of the price of yours. It’s a lot.[/quote]
A car with zero sportiness isn’t a HotHatch. Also stop commenting on the price because my car was one of the few priced at optimum price.
Sure, I’m just going to make an algoridm out of my a** just to show him. Do you think that is reasonable? If Der_Bayer gave me his I would work on improving it. But I’m not even going to ask that because it’s disrespectful to his work. But I’m also not starting my own one because it would take alot of time to get the same result plus the time to improve it. So I criticise what I think is wrong hoping it gets fixed. But next time you write a post in the suggestion area of the forum I’ll try to remember to tell you to go do your own game if you think X feature is wrong.
I made 6 cars for this and I picked the one on the broken group…
[quote=“GenJeFT”]
Nor did I say it was. The two I had in mind for that category would be NormanVauxhall and Kubboz. You and I were targeting the same audience, and priced very similarly. With very different results. You must have had a pretty large factory to be able to produce that many cars. I kept mine small, which was a mistake on my part (and a smart move on yours). So it could be skewed a bit. If other competitors such as myself had made more cars (I sold every car I could make), your numbers may have gone down and someone else’s up. Would it have been enough to alter overall rankings? Possibly. only 20k+ more cars on my part would have put me in 10th for market share. Would it have been enough to alter top-3 in any specific category? Also possible, but less likely.
The key in this scenario to getting overall market share is to make an easy to produce car, build the living daylights out of it in the largest factory you can find, and underprice it somewhat. Doesn’t mean you’re in the running for most profit, and are almost guaranteed not to get most profit per car. (Note… this is for OVERALL market share. Specific model market share is different, obviously)
I’m looking forward to the next challenge. I hope PMP is too. He’s a smart guy with lots of ideas who got an absolutely frustrating result from this challenge. The next challenge will be different, and who knows what the results will be. But I love the competition and seeing how everyone approaches it.
[quote=“VicVictory”]
Nor did I say it was. The two I had in mind for that category would be NormanVauxhall and Kubboz. You and I were targeting the same audience, and priced very similarly. With very different results. You must have had a pretty large factory to be able to produce that many cars. I kept mine small, which was a mistake on my part (and a smart move on yours).
The key in this scenario to getting overall market share is to make an easy to produce car, build the living daylights out of it in the largest factory you can find, and underprice it somewhat. Doesn’t mean you’re in the running for most profit, and are almost guaranteed not to get most profit per car. (Note… this is for OVERALL market share. Specific model market share is different, obviously)
I’m looking forward to the next challenge. I hope PMP is too. He’s a smart guy with lots of ideas who got an absolutely frustrating result from this challenge. The next challenge will be different, and who knows what the results will be. But I love the competition and seeing how everyone approaches it.[/quote]
True, I am actually surprised that Kubboz did not do better then he did. I guess the fact its about 1k more expensive then Normans car made all the difference as there was not enough bang for peoples bucks. But yea, you have to make a car that is fairly easy to make (mine takes about 40 more hours then Normans) and throw up the largest factory you can. The large factory does two things, you can push the price down, and make a LOT of cars. But that strategy relies on 1 thing. Enough people to join so that Der Bayer inflates the number of customers so that you can actually sell every car. Otherwise you can go bankrupt.
I do agree with you all that a car with 0 sportiness isn’t a hot hatch. But the majority of the buyers took a car which suited their needs best. And those were somewhat balanced and cheap cars because the somewhat cheap sporty cars were only sporty and nothing else. If a car close to Normans stats with some sportiness would have been built, the buyers would have rushed for it. But there wasn’t.
@ PMP1337: It’s all about how you give your feedback. Constructive criticism, not “I feel cheated”.
To give you all a better feeling of what happened:
The sales model calculates two scores: one for the stats and one for the costs. The cost score is deducted from the stat score. If the result is below 0, no one will buy the car. That’s why no Hot Hatch buyer bought a supercar. The other way around (don’t sell cheap cars in expensive segments) is not that easy, at least not with hard-coded limits. But still, how cares about 10 cheap sold cars in the supercar category?
This is the scoring algorithm:
stats_score = sum((statsnormalized_stat_weight)^1.8)
cost_score = 2.7 * (10 * exp(1price_weight+2)^1.5*(price/10000)^0.45))^0.075 +
2.7 * (10 * exp(1*(annualcost_weight+2)^1.5*(5*annualcost/10000)^0.45))^0.075
score = (max(0, stats_score - cost_score))^2
In the case Norman vs. PMP1337 in the hot hatch category it worked out this way:
stats_score: N: 87.0; P: 109.6 (PMPs car is the better hot hatch)
cost_score: N: 37.5; P: 74.7 (Norman’s car is muuuuch cheaper)
difference: N: 49.5; P: 34.9
difference squared: N: 2447, P: 1214 -> Norman sells double the cars.
If PMPs car would have been $2-3 k cheaper, the world would have looked different. Just in case someone wants to ask: the squaring in the end is necessary to bring some diversity into the buyer’s choice. Else it would be a big mess where everyone would sell some cars and no one stands out. Nothing you can see in the real world.
I will provide you with an excel sheet (or similar) next round where you can put in your car’s stats and see how it will be scored. This way you can optimise your car for a target group better and the whole process is a bit more understandable.
Still I will not change the “no hard limits” policy. The best hat hatch is not a car with 0 sportiness in general. It’s just that no one built the best car, so the buyers looked for something else.
I aimed fully at the hot hatch category as well, but compared to the competition I just build an overpriced shit car xD so no wonder no hot hatch buyer wanted it. Looking at the strong competition, I as a buyer on a budget looking for a sporty compact would probably have turned down the small sportscar variants too, they just gave too little extra for the premium price compared to the masters of the compact category.
Although, an interesting thought here is: do the earlier years make building sporty cars more difficult? The earlier the year the more you have to invest or sacrifice in order to get any sportiness, while tameness is a bit more difficult too, but much less so. When I look at your calculations that seems to be mirroring that, nothing wrong with it per se, it just makes your (true) statement of:
rather unrealistic because of the current game mechanics making it difficult to get the sportiness values up. I think a pretty good comparison is my car, I basically took a cheap compact and made the appropriate sacrifices + put a more sporty engine in it. The result is a much worse car overall (not YOUR problem), which got a mind-boggling ~13 sportiness for major losses in all other stat categories. This overall is very instructive for the car desirability calculations that have to be put into the game. Also, I think it illustrates nicely why the term Hot Hatch is a pretty modern one, not from '65. 
I agree with the term Hot Hatch being a modern one. Seems like it just wasn’t easily attainable to make a proper hot hatch back then as this now has pointed out.
As for the costs: 5 times annual costs into the calculation! In the real world most customers wouldn’t have such a good idea about them, but I better try taking servicing fees more serious from now on. (although wouldn’t we earn a bit from those as well? 
Too complex to implement for a simple challenge though, probably)
“Annual service cost” is one that kind of gets me all throughout this game. I understand it’s just a percentage of the total cost based on reliability, but even in 2010 dollars it seems… well, insane.
In the 4th year we owned my wife’s Tucson (which was 2009, so $$ wise it’s very close), we spent $800 on its 60k service (timing belts and whatnot), and about $180 on oil changes. That’s it. $980. I’m sure if I recreated the car in-game its annual service cost would be somewhere in the $2k+ range (I have yet to make a car that has less than $1000/yr costs in this game). Well, if you calculate fuel, then yeah, our annual cost for that car was over $2k. But that was the year that we spent far and away more on maintenance than any other year. If a major service or tire replacement weren’t due, we were spending about $200/yr on maintenance. And Der Bayer’s simulation calculates fuel cost separately from annual service cost.
So until that mechanic is better explained (to me) as to what all it entails, I will shake my head, smile, and try to accept it. And if I can’t accept it, I will ignore it as much as I can afford to.
And to answer the sportiness question in 1965: Yes, it might just be that the year is very difficult to balance sportiness with everything else important to a hot hatch buyer. I just did a quick reconfigure of my entry, changing the suspension to double wishbone all over, throwing a 175 HP small v8 in it (was a 97 HP six before), upping it to 5 speeds, closing the gear ratios, giving it meatier tires, and popping the rear brake diameter up a bit. I managed to get 12 sportiness (from 6). 45 tameness. comfort, prestige, and safety all in the low 20’s. My production units went up by 20 man-hours and my material cost nearly doubled. And changing that small V8 out for my 5.6L from my full size cars actually made every stat (except safety and prestige) tank.
I tried to make a GT and it looks like I made a Supercar and Trackday car instead, which I rejected in the first place (well, I was hoping I could get a bit of the supercar market, but not aiming at it) 
I was a lot more conservative on the quality this time because of the changes but it looks like I missed my target. By the looks of it, I should have just increased the price substantially (btw original planned price was 200 000) and make a MUCH smaller factory.
Edit: Looking at the price analysis, I chose the WORST possible price point for my car
Still, best profit per car, I just love making cheaper cars than competitors and selling them for more, I just hope my customers wont feel ripped off and boycott me in the next challenge
Great challenge, so much depends on what the competition brings!
[quote=“Killrob”]I aimed fully at the hot hatch category as well, but compared to the competition I just build an overpriced shit car xD so no wonder no hot hatch buyer wanted it. Looking at the strong competition, I as a buyer on a budget looking for a sporty compact would probably have turned down the small sportscar variants too, they just gave too little extra for the premium price compared to the masters of the compact category.
Although, an interesting thought here is: do the earlier years make building sporty cars more difficult? The earlier the year the more you have to invest or sacrifice in order to get any sportiness, while tameness is a bit more difficult too, but much less so. When I look at your calculations that seems to be mirroring that, nothing wrong with it per se, it just makes your (true) statement of:
[quote]If a car close to Normans stats with some sportiness would have been built, the buyers would have rushed for it. But there wasn’t.[/quote]
rather unrealistic because of the current game mechanics making it difficult to get the sportiness values up. I think a pretty good comparison is my car, I basically took a cheap compact and made the appropriate sacrifices + put a more sporty engine in it. The result is a much worse car overall (not YOUR problem), which got a mind-boggling ~13 sportiness for major losses in all other stat categories. This overall is very instructive for the car desirability calculations that have to be put into the game. Also, I think it illustrates nicely why the term Hot Hatch is a pretty modern one, not from '65. [/quote]
I was aiming for the $compact and hot hatch categories. My car was paterned after a low-end Alfa Giulia model. I eneded up scorring most in the $mid-size and $$Mid-size. Go figure.
I aimed for and got most of my sales in ££mid .
Although I was surprised at some of the overspill into other cats.
Given that I sold all my stock I have a question about the full game. Would limited availability increase a car’s desirability? That way you can charge more for it.
Der_Bayer, because so much depends on the price and you simulate 5 years, would it be possible to adjust prices for example after the first year of sales? Close the competition, make a simulation for one year and post partial results (you dont have to post all, maybe only finance overview or just total sales), leave 24hours for price adjustments of participants and then simulate the remaining 4 years? Or make automatic price adjustments per year (but only in x% increments), but that would probably be pretty hard because not everyone is here to maximize profit, some want to maximize marketshare. But nobody in their right mind would sit on overpriced cars without any discount, or keep selling cars cheap if they cant keep up in production
But I understand if its too much hassle for you and you want to keep it simple&easy to manage. I would settle for more frequent challenges
Ack, my car was genuinely terribad. I would not have made a profit no matter what price I set. LOL
Hmmmm… I cannot get the costs score formula to give the same result as it should.
I think that we should look to this: A hot-hatch buyer likes more sportiness, but also prefers some tameness and safety. But if your car has much of these last two, your car will reach another group who gives more attention to that stats. And you will up ending selling cars to those 2 or 3 groups. If you’re able to moderate your car specs to reach the necessary amount of stats for a determined group (which is difficult) you simply hit the target group successfully. Even if we sold some cars to other groups, the majority focus is in the aimed group, based on the well balanced stats. In certain groups, we need to reduce, per example, the security stats.
If we pay attention to marketshare sheet, we’re able to see the colours of the stats that we hit. Compare that with your car and with the target groups. Compare the colours. I.e. > 50 is green… < 50 yellow…and so on… It allows us to focus in one group or more.
VW doesn’t sell a Golf GTI just for hot hatch lovers…
I think the best approach is not aiming at a specific target group, but instead try to find overlap points. Check my car for example, wasn’t great in a particular segment, but it was good across the board, with sales in every segment except for $compact. And it ended up in second place.