Colonist Expansion Profitable? - A Real Example

[Stalker0]

So I decided that arduously recording one city wasn't enough pain, so I did it again! I took the same game as my Pioneer game (Found here: Pioneer Expansion Profitable? - A Real Example | CivFanatics Forums ), and added on a Colonist made city to see how it did. I put most of metholody in the other thread, so look there if you want more details.

Behold Kazan, my 11th city!

Now one thing I did do differently based on feedback, I tried feeding the colony a hammer ITR right at the beginning. This TR effectively "costs" me gold, science, and culture every round, so I add those numbers in as costs for running the colony.

Spoiler :

The Build Order

Spoiler :

From Pioneer

• Forge
• Council
• Library
• Market
• Barracks
• Granary
• Monument
• Shrine
• Well

From Colonist

• Windmill
• Workshop
• Amphitheater
• Temple
• Aqueduct
• Harbor
• Lighthouse

Actual Buildings (in Order, all buildings invested)

1. Walls
2. Bank
3. Chancery
4. Castle
5. Herbalist
6. Arsenal
7. Bath
8. Customs House
9. Arena
10. University
11. Armory
12. Opera House
13. Grocer
14. Public School
15. Constabulary
16. Caravansary
17. Stables
18. Circus
19. Zoo
20. Garden
21. Wire Service
22. Civilized Jewelers Office

The Results
Colony Ran from Turns 297 - 390 (93 turns). My totals (after costs, aka pure profit)

Total Colony Science: 6,387.35
Total Colony Culture: -933.05
Total Colony Gold: 1,964.84
Total Colony Faith: 431.00

Total Science: 151,750
Total Colony Science (before costs): 10,742
Total Science (Before Colony): 145,563 (note my inflation factor is now 3.23% for an 11th city, instead of 3.33% from my last run).
% Colony Science Increase: 4.25%

Total Culture: 208,589
Total Colony Culture (before costs): 5,544
Total Culture (Before Colony): 209,603
% Colony Culture Decrease: -.48%

Overall Colony Science+Culture Increase: 3.77%
Overall Pioneer Science+Culture Increase: 3.2% (from previous thread)

 

[Stalker0]

So now to an interesting question we can actually answer, the longer the game goes, is my colony % contribution going up or down? Aka is having the colony gives me greater returns the longer the game runs.

So with our current data, we can stop the game at various points and see how well the colony did.

Turn | % contribution

341: 5.23%
360: 5.14%
390: 3.77%

So we see an interesting trend, the longer the game goes, the less "important" the colony is to our overall success. Aka our more developed cities contribute "even more", and overtake the extra yields that our colony provides. This I think flies in the face of our default assumption, which would be that a longer game would give the colony more time to be at max power, and would in theory then contribute more overall....but that's not what the numbers show.

 

[CrazyG]

Turn | % contribution

341: 5.23%
360: 5.14%
390: 3.77%

I wonder what these numbers would look like for all cities in the empire. I'm guessing that this colony is low on culture since it can't have a guild but science seems odd at a glance. This strategy seems to have 0 factors rewarding pure tall, right?

 

[Stalker0]

I wonder what these numbers would look like for all cities in the empire. I'm guessing that this colony is low on culture since it can't have a guild but science seems odd at a glance. This strategy seems to have 0 factors rewarding pure tall, right?

Part of this maybe the fact I am playing Industry. As time goes on, more of my core cities start running science process a higher % of the time, so their science contributions go up (I also ran art processes for a while until I got my first tier 3 policy, then switched hard over to science). So that may be a more industry specific finding.

And yeah done of my religion or policy settings are tall favored.

 

[AndreyK]

I think this analyse goes to concrete city and map, because your yields per turn still depend on tiles that you work and resources available.

 

[Stalker0]

I think this analyse goes to concrete city and map, because your yields per turn still depend on tiles that you work and resources available.

Apologies, but I'm not really sure what you are trying to say.

 

[AndreyK]

Well, I think that city yields come also from tiles that city works. For example if you have jungle tiles those give you science, forests and jungle give you culture with zoo (especially with Spain bullring). There are lots of variations.
There could be resources that give you culture, science and gold. So I think that give big difference to city protectiveness on all yields.

 

[Stalker0]

Well, I think that city yields come also from tiles that city works. For example if you have jungle tiles those give you science, forests and jungle give you culture with zoo (especially with Spain bullring). There are lots of variations.
There could be resources that give you culture, science and gold. So I think that give big difference to city protectiveness on all yields.

ah I understand now thank you. So one of the wonderful things that this data offers me, is I can estimate certain "what if" scenarios.

Moar Wood! What if Scenario
So for example, lets say that my city had an additional 3 forests (so 6 in total), and I had been working those forests since the zoo was built. Now the problem here is population, I would in theory have to remove pop off specialists to work the forests. Since my specialists give culture and science....that would decrease my key yields not increase them. But lets go for a high end estimate, lets just say we had the extra 3 pop magically granted to us, so we can see what it looks like.

So 3 forests (in a triangle of course!), grant 3 culture and 27 hammers. Lets add in some good bonus multipliers and make it 4 culture and 32 hammers. 32 hammers -> 8 science through processes. Now again I wasn't running processes the whole time since I made the zoo, but I would in theory have made the buildings a little quicker as well (not that much, many buildings were already finishing in a single turn), but it would have been something. So again, this is a high end estimate, so lets just say I got to turn all of those hammers into lovely science.

So I ran the zoo for 71 turns, yielding me 284 culture and 568 science.

The end result (which we must remember is an upper bound, all of our assumptions are giving us yields higher than is realistic, but just to showcase what could be possible).

Our contribution goes from 3.77% -> 4.29%

 

[Stalker0]

Here is another great what-if scenario, hehe you want to talk about high end estimates, here is the biggest.

Ultimate High End What If

Lets assume for a moment that when we made the city, it was already 100% productive right from the start, heck lets go better....lets say that every turn the city was in existence it made the HIGHEST science and culture that it ever makes....ever, every single turn. Heck I'm giving it access to buildings it doesn't even have yet in the early turns. But we will also say that I still get all of the instant science and culture credit for making the buildings I made....so no losses added, its all gravy baby!

I hope we can all agree that this is a crazy scenario, absolutely unrealistic and high end. Aka no colony would ever reach this level of yield deliverance.

The result? It changes the % improvement from 3.77% -> 7.72%.

So I think we can very safely say that no colonist made city, no matter how cool, how optimized, no matter your policies or religion or special civ bonus, is going to break 7% (heck probably 6% is completely impossible but I would need more math to say that).

5% Scenario
Lets try another. Lets say we wanted to hit 5%, a nice solid number. Lets say I could magically add +X SPT to my colony, right from the day it was founded. How much SPT would it take to get to that 5%?

Answer: ~+19.5 SPT, so effectively you would need to found the colony with 1.5 extra academies to make up the difference.

+1% Scenario
And lastly, lets just say I wanted to increase the % by 1. Aka how many SPT would I need to add every turn to increase the rate by 1%?

Answers: ~+15.9 SPT, and that scales linearly. So 31.8 SPT is 2%, etc.

Authority instead of Progress
Ok this one is a little trickier to estimate, but still doable. So what if we had founded our colony under an Authority + Industry regime instead of Progress + Industry?

So we gain a +2 CPT from the garrison (lets just assume some really good culture bonuses and make that a straight up +3 CPT).

Authority gains +4 HPT over Progress, and has tribute to generate even more hammers. Meanwhile Progress has +10% hammers when building. To keep things simple, I will assume the +4 HPT and +10% hammers wash out, and I will convert the tribute hammer bonus into science on a 4 H to 1 S ratio (for science process). I am actually giving Tribute more due here than I should (I am giving it the same border growths even though progress has more culture), but lets go for it and see. Obviously Authority loses the building instant bonuses but does gain a big bonus when the colony is founded.

So what would the % increase look like here as an estimate?

3.77% drops to 3.2%.

 

[Stalker0]

Updated OP with datasheet, its cleaned up and ready for review.


Another idea I tried, was to apply a light "yield discount" to see how it would adjust matters. We all know that yields now > yields later in Civ due to the snowball mechanics of the game. This means that 100 science on Turn 300 is worth less than the same 100 science gained on Turn 200.

My discount was a linear one (aka it increases by the same amount each turn) and is 1/450. That means that on Turn 1 yields have full value, on Turn 225 are halved, and on Turn 450 they are "worthless" (aka the game is over).

This is not perfect by any means, a true discount strategy is likely to be extremely complex and probably not linear (and would vary yield to yield), but we at least try to account for it a little bit.

Doing that adjusted my % Increase from 3.77 -> 4.04%. That's interesting to me, but it reinforces the trend we saw earlier where my % increase was going down over time. Effectively the colony is most productive to the civ in its middle turns, after its investment, but before the rest of the civ finishes their buildings and start going pure science themselves.

The discount magnifies those middle turns a bit, as the science and culture gained during that time is a bit more valuable than the yields gained towards the end of the game. Now technically the early costs of the colony are also increased by this method (as they are the earliest turns they receive the least discount), but it seems that the middle productive turns are the most important.

 

[Stalker0]

Another datapoint that I think is interesting is....how much of the colonies science and culture (as well as the entire civ) was generated with raw yield versus instant yield? Aka are instant yields just an add on, or a core part of your science engine?

This run should indicate a higher value in general, as we are using Progress + Industry, both of which provide a large amount of instant yields on science/culture. So this could represent an upper bound in general, or close to it.

For the 92 turns in question:

Science
Colony: 15% instant yield / 85% raw
Civ: 40% instant / 60 % raw

Culture
Colony: 19% instant / 81% raw
Civ: 39% instant / 61% raw

So the civ numbers especially are pretty high on instant yields. Part of the reason why is that this is the "GP bulb time". All of your GS and GW are getting bulbed for big science and culture, which is likely pushing these numbers. But it does show that instant yields cannot be ignored in any real calculation, they are a core part of your yield engine.

 

[Stalker0]

The Power of Puppets (well....actually the lameness of puppets)

Another what if. What if the colony did not increase our science %? Like a puppet city, all my yields are pure profit. So how much would a puppet have to contribute to hit that same increase. To keep this simple, I will assume no culture production or loss (so culture % is 0). So our science % has to be 3.77% to match the original colony. Note that it takes more culture to move the % needle than science, so I'm being kind to the puppet here. If I had instead chosen culture and kept science at 0, the puppet would have an even greater hill to climb.

Answer is 5,530 science, or 60 SPT. If we use the normal puppet math (80% of science is lost), that's actually 300 SPT! If we assume that 50% is done through instant yields (which are not subject to the puppet penalty as far as I know).... than that would be 180 SPT. And then the best scenario, 50% is instant yields AND only 60% loss (imperalism). That would be 105 SPT.

This is telling, because even if we assume a puppet was able to generate the same infrastructure just as quickly (which we know is patently false), a puppet simply cannot generate that amount of science (those are more like capital numbers!) So I think its reasonable to say that regardless of whether you think the colony is generating a "good" amount of science overall, a puppet would actually have been much less yield efficient.

 

[CrazyG]

So the civ numbers especially are pretty high on instant yields. Part of the reason why is that this is the "GP bulb time". All of your GS and GW are getting bulbed for big science and culture, which is likely pushing these numbers. But it does show that instant yields cannot be ignored in any real calculation, they are a core part of your yield engine.

If possible you should post these numbers in three groups, Raw, Instant, and Great Person, just because great person yields depend on the first two.