Imperial census

[Carlobrand]

I completed a little pet project: an Excel-based (well, OpenOffice-based) census of the Imperium, based on the data in the Wiki and TravellerMap. And:

(drum roll)

The Imperium consists of 8926 worlds with a total population of just a bit over 15 trillion enumerated sapients (as opposed to those who for one reason or another didn't get counted), not counting the 31 Vegan worlds.

(There are actually quite a lot of Vegans. Not clear how many humans or other species are living on Vegan-majority worlds, but there are 174 billion enumerated sapients among the 31 Vegan worlds, of which 20 billion are on clearly identified human-majority worlds - something about the Vegans not liking high gravity. For a minor race, the Vegans get around.)

Economy-wise - well, it depends on which model you use. I favor the Striker model myself: that yields a total gross world product of 265.7 billion megacredits for the Imperium - I guess that's 265.7 petacredits. I had to look that one up, never had to use it before. WIlling to work with other models if they can be adapted to the available data.

One caveat is that I understand there's some question as to the method for determining the secondary population code. If that gets changed, I need to change my numbers.

 

[MarcPolo]

Dandy.
Might you share the spreadsheet or detail your economic calculations?
Thanks too.

 

[DonM]

As the guy who hands the data to travellermap and the wiki, what's the problem with the secondary pop code?

And if you've got questions/findings/suggestions, drop me a note at don.mckinney@gmail.com.

 

[Carlobrand]

Dandy.
Might you share the spreadsheet or detail your economic calculations?
Thanks too.

Be glad to, but there are 29 sheets - one per sector - 'cause the program crashed when I tried to do them all on one file. I'll put them up on the weekend, if that's okay.

As the guy who hands the data to travellermap and the wiki, what's the problem with the secondary pop code?

And if you've got questions/findings/suggestions, drop me a note at don.mckinney@gmail.com.

I inquired about a discrepancy between the numbers I was getting from the data sheets and the numbers showing up on the wiki sector write-ups for sector population. I got a response back that the process they were using to calculate the sector populations used a difference in interpretation of the population modifier than was customary.

http://www.travellerrpg.com/CotI/Discuss/showthread.php?t=32431

 

[DonM]

I'll have to ask Thomas about how the wiki determines population.

For my analysis, I use an Excel workbook, with one page of explanations (to remind me what I was doing), one page of reference tables, one page for analysis summaries, and then each sector has its own page.

 

[rancke]

I'll have to ask Thomas about how the wiki determines population.

For my analysis, I use an Excel workbook, with one page of explanations (to remind me what I was doing), one page of reference tables, one page for analysis summaries, and then each sector has its own page.

Have you and Marc decided how to handle non-mainworld populations?


Hans

 

[Carlobrand]

...For my analysis, I use an Excel workbook, with one page of explanations (to remind me what I was doing), one page of reference tables, one page for analysis summaries, and then each sector has its own page.

I wish I could afford Excel. I loved that program when I had access to it, but I'm working in OpenOffice now, which has the advantage of being free and being able to read all my old Excel files - very important, since I have a lot of them - but the disadvantage of not being as user-friendly or as robust. I tried building one file with all the sectors and it started crashing after about 10 pages.

So, when I upload this, do you guys want it as OpenOffice ODS or as an Excel extension?

 

[tjoneslo]

I'll have to ask Thomas about how the wiki determines population.

Two questions:

1) Did you, as part of the T5SS updates, regenerate the population multiplier (P in the PBG code) according to the flat number generation as described in the T5 Core Book?

2) Echoing Hans' question, are there any rules for calculating the population of secondary worlds, transitory populations, or other non-main-world populations?

 

[DonM]

Two questions:

1) Did you, as part of the T5SS updates, regenerate the population multiplier (P in the PBG code) according to the flat number generation as described in the T5 Core Book?

2) Echoing Hans' question, are there any rules for calculating the population of secondary worlds, transitory populations, or other non-main-world populations?

The existing "P" was declared canon. Is there a problem with the existing P numbers? And don't point at distribution. A ton of those numbers were deliberately massaged by the Old Ones.

For Hans' question, I'm still not getting the issue. The rules for generating secondary worlds is in the book, and isn't all that different from previous rules for non-mainworld pops. What is the issue there? Are you guys looking for a formula for non-mainworld pops per system so you have that number without generating the other secondary worlds in a system?

And what's a transitory population? The number of people in jump space during a average week across the Imperium?

 

[tjoneslo]

The existing "P" was declared canon. Is there a problem with the existing P numbers? And don't point at distribution. A ton of those numbers were deliberately massaged by the Old Ones.

There is the distribution described by rules used to generate the values, and there is the distribution of the values as presented in the data. In order to answer the question about the distribution of the P values valid, I'll need to do the statistical analysis.

 

[BeRKA]

Benford’s law

There is the distribution described by rules used to generate the values, and there is the distribution of the values as presented in the data. In order to answer the question about the distribution of the P values valid, I'll need to do the statistical analysis.

Benford’s law is what you should use to check the P values:

 

[aramis]

Benford’s law is what you should use to check the P values:

Traveller massively fails benford's rule.

 

[rancke]

Benford’s law is what you should use to check the P values:

Only if you assume that planetary populations start at level 1 and grow from there. But an outpost starts at whatever population is required to fulfil its purpose and may never grow bigger. Does crew sizes on oil platforms conform to Benford's Law?

Populations of other worlds may follow Benford's law -- until they decide to implement a zero-growth policy (if they do). There may be breakpoints where population growth rates tend to change.


Hans

 

[Carlobrand]

I don't know who this Benford guy is, but I sure didn't vote for him.

Mathematically the law makes sense for any system undergoing predictable growth. However, as Rancke points out, a Traveller system may not undergo growth the same way another system might. A corporate-owned purpose-driven settlement is not going to grow beyond that supported by the jobs assigned by the corporation - assuming they allow families, any kids not afforded jobs will have to emigrate. A settlement undergoing a "gold boom" may grow from a few hundred to tens of thousands in a matter of months as people immigrate in to take advantage of the boom.

For very large population worlds, Benford's Law works fine, but below a certain population it's going to depend on very local circumstances and will look a lot more like how towns grow - or don't grow - than how planetary populations grow.

 

[tjoneslo]

I would strongly recommend reading the Wikipedia article.

The law has nothing to do with population growth, or lack thereof, but is instead an observation of the distribution of the first digit of numbers. It applies in to a wide variety of situations.

More importantly, it applies only when you are comparing a large number of samples. In the cases I'm working with somewhere between 12,880 (the number of worlds in the official T5 Second Survey) and 47,000 (the number of worlds in charted space). So you can claim that world X does not fit into Benford's law, or a small number of worlds don't fit. It won't impact the overall analysis.

 

[Carlobrand]

I would strongly recommend reading the Wikipedia article.

The law has nothing to do with population growth, or lack thereof, but is instead an observation of the distribution of the first digit of numbers. It applies in to a wide variety of situations.

More importantly, it applies only when you are comparing a large number of samples. In the cases I'm working with somewhere between 12,880 (the number of worlds in the official T5 Second Survey) and 47,000 (the number of worlds in charted space). So you can claim that world X does not fit into Benford's law, or a small number of worlds don't fit. It won't impact the overall analysis.

I did. The problem is, at least as Traveller generates worlds, a rather large number of worlds may not fit. The base population code is a 2D6-2 roll that represents a factor of 10. Means the average roll is 5: hundreds of thousands. 15 in 36 worlds will have populations in the tens of thousands or lower, and all of those are subject to extraneous factors that swamp the standard analysis. Someone shouts, "GOLD!" and the population doubles or triples in a couple months. One good tsunami and the population is all but extinct.

There's no question that the straight D10 is a statistical obscenity, but Traveller has a whole lot of small population worlds that will behave more like small towns and villages population-wise than like planets. While they will TEND to follow Benford's law, they will ALSO be far more affected by random variables not accounted for in any rules, and these variables will pretty much swamp any numerical analysis. So, there's no real pressure to get those worlds perfect - 'cause there's no real way to get them perfect.

 

[aramis]

I did. The problem is, at least as Traveller generates worlds, a rather large number of worlds may not fit. The base population code is a 2D6-2 roll that represents a factor of 10. Means the average roll is 5: hundreds of thousands. 15 in 36 worlds will have populations in the tens of thousands or lower, and all of those are subject to extraneous factors that swamp the standard analysis. Someone shouts, "GOLD!" and the population doubles or triples in a couple months. One good tsunami and the population is all but extinct.

There's no question that the straight D10 is a statistical obscenity, but Traveller has a whole lot of small population worlds that will behave more like small towns and villages population-wise than like planets. While they will TEND to follow Benford's law, they will ALSO be far more affected by random variables not accounted for in any rules, and these variables will pretty much swamp any numerical analysis. So, there's no real pressure to get those worlds perfect - 'cause there's no real way to get them perfect.

I've had fun using "roll 2d9, keep lowest" for PopMults. It generates a relatively reasonable approximation of Benford's law.

 

[Ishmael]

I just generate 3 decimal places on the end of the pop number; for example: 6.375 .
the regular popmod, as the rules use it, then becomes 10^.375, from the example.
To increase the pop by 1%, I just add .004321 ( depending on how many decimal places you care to keep ) to the pop number.

and it is consistent with benford's law

 

[tjoneslo]

There's no question that the straight D10 is a statistical obscenity, but Traveller has a whole lot of small population worlds that will behave more like small towns and villages population-wise than like planets. While they will TEND to follow Benford's law, they will ALSO be far more affected by random variables not accounted for in any rules, and these variables will pretty much swamp any numerical analysis. So, there's no real pressure to get those worlds perfect - 'cause there's no real way to get them perfect.

But that's exactly the point. If you are assuming a dynamic universe where several worlds are changing at the same time, each world change may be effectively random but the overall statistics don't change.

Benford's law isn't valid for one world, or even a small number. Even the wiki article points out the law isn't valid if you don't use a broad enough population. But I'm assuming 47,000+ worlds of charted space , with a population range 9 orders of magnitude.

 

[DonM]

I don't think Benford's Law will work, because we know that the numbers were fudged by the Great Old Ones of Traveller.

Trust me: I'm buried right now in just how different Solomani Rim's stats are from everything else. Old Expanses and Massilia are a little off, but Solomani Rim just blows the doors off.