Price per Piece - An outdated idea?

[nerdsforprez]

1 hour ago, Bartybum said:

Sorry sorry, I meant R^2 closest to 1. Entirely my bad, I got R^2 inversely confused with relative error or whatever the stats term is.

No prob

 

52 minutes ago, Bartybum said:

Okay, I think I understand. Since the data varies wildly, none of it is really consistent.

Well somewhat.  Variance is actually a GOOD thing when looking at the relationship between two variables. In terms of correlation between two things (variables), the more appropriate way to term their relationship is CO-variance.  The amount of variance they share.  If one variable has little variance, or both do, this can severely restrict a correlation.  If both are highly variable, but they vary together - you can still get a high r squared value.  In reality- that is really the only way to do so.  

Edited August 13, 2019 by nerdsforprez

[Bartybum]

@nerdsforprez Do you work with statistics for a living?

[nerdsforprez]

39 minutes ago, Bartybum said:

@nerdsforprez Do you work with statistics for a living?

Answered by PM

[nerdsforprez]

Not to beat a dead horse but I do like this thread (of course I do, its mine ) and more importantly the input from others.

I will try to update it from time to time when new sets present themselves.

The weight for 42099 is finally out (at least on BL).  A paltry (relatively) 2156 grams.  So, not only is the PPP extremely high at 26 cents per piece the Price Per Gram is extremely high as well.  Kinda wrecks the models we have derived so far.  Comes in at twelve cents per gram.  To put this in context, the next highest priced set of the last 6 years (w/o inflation, so actually the band is likely a little narrower) is 8043 at 8 cents per gram.  42030 is 7 cents per gram. 

So, with no cost being paid to a Brand or Trademark, I can only surmise that we really are paying for a premium for the new non-motorized parts and the development for the new Control+ system.    I don't make this statement as an evaluation of any sort, rather just a descriptive one.  One can easily argue that the development of such a new system is very expensive.  I will be interested to see if 42100 shares this cost as well.  If 42100 comes in at a much less PPG, only then will I really wonder what is up with the pricing of this set. 

[suffocation]

23 minutes ago, nerdsforprez said:

If 42100 comes in at a much less PPG, only then will I really wonder what is up with the pricing of this set. 

First off, thanks for starting the thread and doing all the maths

Compared to 42099, 42100 is almost sure to be chea... uh, less extortionate in terms of price per gram:

• less than twice the price
• more than twice as many motors
• more than four times as many parts

Not sure how much the weight of 42099's four Xerion tyres is going to shift things one way or the other.

[nerdsforprez]

3 hours ago, suffocation said:

First off, thanks for starting the thread and doing all the maths

Compared to 42099, 42100 is almost sure to be chea... uh, less extortionate in terms of price per gram:

• less than twice the price
• more than twice as many motors
• more than four times as many parts

Not sure how much the weight of 42099's four Xerion tyres is going to shift things one way or the other.

Agreed. Not to mention 42100 carries the Liebherr name.  Which means that 42099 really is unique price-wise.  So not only is there a premium for Control+ bur b/c of Control+ .....AND it was the first one out? (Just a guess)

[aol000xw]

So did you guys reach a conclusion?

A price per gram that could be used as reference?

 

[nerdsforprez]

Of course it can be used as a reference.  In fact, it is more accurate as a reference than PPP is.  The question was if PPP was completely null and void. 

Based on the input from others, I don't think it is.  Although PPG is more accurate, it is more tedious to calculate (weight not as easily accessible as piece count) and still accounts for ALOT of price of a set.  Additionally, as pointed out by others. piece count my add to the building experience where PPG likely will not. 

PPG is even more valuable in sets with tons of PF or other motors, etc.  Or... as mentioned in the SW forum..... sets that are anticipated to carry many large, heavy pieces.  I think the press release thread for the new UCS Star Destroyer

 

 

 

Is a great example.  At 699.99 USD and only just over 4500 pieces it comes to an abysmal ~ 15 cents per piece.  However, many pieces are large, technic beams or panels as well as large regular system panels.... so lots of ABS.  We don't actually have a weight for the set but I bet that it will be much more accurate in its relationship to price than the piece count is. 

 

 

 

Edited September 6, 2019 by nerdsforprez

[icm]

The question was not if PPG can be used as a reference, but what the reference number should be.

[Mechbuilds]

Soo the 42081 costs 150 euros and weighs 2,35 Kg's. 

When we divide 150 by 2350 we get 0.06€ per gram? 

The new 42099 weighs 2.304 kg and costs 200€ 

So that's 0.08 per gram. 


So the project volvo is a better bang for the buck? 

[Bartybum]

6 hours ago, Mechbuilds said:

Soo the 42081 costs 150 euros and weighs 2,35 Kg's. 

When we divide 150 by 2350 we get 0.06€ per gram? 

The new 42099 weighs 2.304 kg and costs 200€ 

So that's 0.08 per gram. 


So the project volvo is a better bang for the buck? 

When looking purely at PPG then yes, but really it’s gonna be much more complex when you start considering functions, special elements, appearance, etc.

So in other words yes, but actually no