I Scraped a Bunch of Alliance City Building Data (Sponsored By DNN)

[hidude45454]

Sponsored by DNN: https://discord.gg/MDYdBkzFh4

Individual cities built (tab 1): https://docs.google.com/spreadsheets/d/1D_BAnIP854eDHwnrWhDLaf6b3Ho0iy3r5TltFxSD27s/edit#gid=2048071689

Alliance information (tab 2): https://docs.google.com/spreadsheets/d/1D_BAnIP854eDHwnrWhDLaf6b3Ho0iy3r5TltFxSD27s/edit#gid=165451901

Alliance percent increase (tab 3): https://docs.google.com/spreadsheets/d/1D_BAnIP854eDHwnrWhDLaf6b3Ho0iy3r5TltFxSD27s/edit#gid=180076824

Alliance information by tier (tab 4): https://docs.google.com/spreadsheets/d/1D_BAnIP854eDHwnrWhDLaf6b3Ho0iy3r5TltFxSD27s/edit#gid=1687370431

Individual alliance matchups (tab 5): https://docs.google.com/spreadsheets/d/1D_BAnIP854eDHwnrWhDLaf6b3Ho0iy3r5TltFxSD27s/edit#gid=2030059055

Some graphs (tab 6): https://docs.google.com/spreadsheets/d/1D_BAnIP854eDHwnrWhDLaf6b3Ho0iy3r5TltFxSD27s/edit#gid=2047832592

Everything on the sheet is pretty intuitive so I don't think I need to explain anything, but a few caveats --

0. This was pulled like a week ago for anyone wondering why some things look kind of different.

1. City cost is assumed to be using optimal costs (best projects, policy, etc.) so that I could estimate a lower bound instead of a higher bound.

2. This doesn't factor if players switched alliances -- for example, if someone left TKR to Cataclysm in the past 6 months, the cost of the cities they built in TKR will be added to the Cataclysm data point regardless. (No convenient way to track this and it doesn't make a huge difference regardless)

3. Obviously this is not just from alliance grants and loans, but also from a player's own nation revenue.

Econ is by far my worst area, so I can't really comment much on this data, but I do welcome other people to look it over and point out anything interesting they might notice. If I can say one thing though, it's that the data is a nice way of quantifying some things that most people already know about city-building in PnW. For example, whales spend more building cities (duh), alliances with more cities also build more cities (also duh), and alliances with more members generally build more cities (also also duh). I guess that also gives a way to answer questions like, why does Rose build so many more cities than everyone else? They have a large number of members, which leads to many more sources of alliance bank revenue from members (taxes, resources from members quitting, new player raiding, communal warchests, etc.) This makes it quicker to build a city for an individual player, and then revenue from that new city goes into funding other players, and so on. And Rose is also reasonably high tiered, so members are able to make more revenue and build more cities/contribute to higher tax revenue. Not pictured from the data -- Rose also focuses city-building more than most alliances do and probably also puts members in more debt to build cities than most alliances do as a result (and also since they can afford it). (Addendum: also very likely this is the case with TFP, in case anyone was surprised to see them so high.) Following that observation to its logical conclusion, I am tempted to ask stupid questions like, would merging your entire sphere into one alliance Treasure Island-style and then taxing them all at 100% NPO-style be the fastest way to build cities? I do wonder if this is what Alpha was complaining about before they got banned as well. Feel free to answer that if you'd like xP

And as always, feel free to suggest additional things to add to the sheet!

Edited April 7, 2022 by hidude45454

[Darth Ataxia]

Thanks for your data work, mr.hidude

[KillzBob]

Did someone say data?

[Canbec]

I do love a good dataset. Thanks for putting this together!

[Indger]

6 hours ago, hidude45454 said:

Sponsored by DNN: https://discord.gg/MDYdBkzFh4

Individual cities built (tab 1): https://docs.google.com/spreadsheets/d/1D_BAnIP854eDHwnrWhDLaf6b3Ho0iy3r5TltFxSD27s/edit#gid=2048071689

Alliance information (tab 2): https://docs.google.com/spreadsheets/d/1D_BAnIP854eDHwnrWhDLaf6b3Ho0iy3r5TltFxSD27s/edit#gid=165451901

Alliance information by tier (tab 3): https://docs.google.com/spreadsheets/d/1D_BAnIP854eDHwnrWhDLaf6b3Ho0iy3r5TltFxSD27s/edit#gid=1687370431

Individual alliance matchups (tab 4): https://docs.google.com/spreadsheets/d/1D_BAnIP854eDHwnrWhDLaf6b3Ho0iy3r5TltFxSD27s/edit#gid=2030059055

Some graphs (tab 5): https://docs.google.com/spreadsheets/d/1D_BAnIP854eDHwnrWhDLaf6b3Ho0iy3r5TltFxSD27s/edit#gid=2047832592

Everything on the sheet is pretty intuitive so I don't think I need to explain anything, but a few caveats --

0. This was pulled like a week ago for anyone wondering why some things look kind of different.

1. City cost is assumed to be using optimal costs (best projects, policy, etc.) so that I could estimate a lower bound instead of a higher bound.

2. This doesn't factor if players switched alliances -- for example, if someone left TKR to Cataclysm in the past 6 months, the cost of the cities they built in TKR will be added to the Cataclysm data point regardless. (No convenient way to track this and it doesn't make a huge difference regardless)

3. Obviously this is not just from alliance grants and loans, but also from a player's own nation revenue.

Econ is by far my worst area, so I can't really comment much on this data, but I do welcome other people to look it over and point out anything interesting they might notice. If I can say one thing though, it's that the data is a nice way of quantifying some things that most people already know about city-building in PnW. For example, whales spend more building cities (duh), alliances with more cities also build more cities (also duh), and alliances with more members generally build more cities (also also duh). I guess that also gives a way to answer questions like, why does Rose build so many more cities than everyone else? They have a large number of members, which leads to many more sources of alliance bank revenue from members (taxes, resources from members quitting, new player raiding, communal warchests, etc.) This makes it quicker to build a city for an individual player, and then revenue from that new city goes into funding other players, and so on. And Rose is also reasonably high tiered, so members are able to make more revenue and build more cities/contribute to higher tax revenue. Not pictured from the data -- Rose also focuses city-building more than most alliances do and probably also puts members in more debt to build cities than most alliances do as a result (and also since they can afford it). (Addendum: also very likely this is the case with TFP, in case anyone was surprised to see them so high.) Following that observation to its logical conclusion, I am tempted to ask stupid questions like, would merging your entire sphere into one alliance Treasure Island-style and then taxing them all at 100% NPO-style be the fastest way to build cities? I do wonder if this is what Alpha was complaining about before they got banned as well. Feel free to answer that if you'd like xP

And as always, feel free to suggest additional things to add to the sheet!

Did you miss me and my alliance in the sheet?

[SweetButPsycho]

Found my nation on the sheet, now I feel shy lmao

[Wendell]

1. The money/person (6mo)vs Members graph is very interesting. I think it could be better represented with a bar graph. Anyway it tells you what kind of econ program they have. Members in a lot of alliances are left obviously left to pay for more of their cities. The outliers are of course top 10 alliances that can afford to pay for cities. But the interesting thing is I see a lot of older high tier, medium size alliances (OWR, ASM, The Wei, UPN, Camelot) to the left of the graph. Have they implemented a city building program at all? Where is their income going? Why aren't the further right?

[Sweeeeet Ronny D]

These numbers are out of date, grumpy has spent over 10 billion dollars the last few days picking up cities. AKA we bought like 3-4 cities the last few days.

1 hour ago, Darth Tryptophan said:

But the interesting thing is I see a lot of older high tier, medium size alliances (OWR, ASM, The Wei, UPN, Camelot) to the left of the graph. Have they implemented a city building program at all? Where is their income going? Why aren't the further right?

This is because they are crappy alliances.

 

Tho to all of you that !@#$ about how its impossible to catch up to Grumpy, this gives you an idea of how long and how hard it is for Grumpy to pick up cities.  You can see out of our guys with over 40 cities, we average about 2 cities per 6 months based on this data.

Edited April 6, 2022 by Sweeeeet Ronny D

[Raphael]

I love stuff like this, thanks Hidude.

[Wendell]

1 hour ago, Sweeeeet Ronny D said:

These numbers are out of date, grumpy has spent over 10 billion dollars the last few days picking up cities. AKA we bought like 3-4 cities the last few days.

This is because they are crappy alliances.

 

Tho to all of you that !@#$ about how its impossible to catch up to Grumpy, this gives you an idea of how long and how hard it is for Grumpy to pick up cities.  You can see out of our guys with over 40 cities, we average about 2 cities per 6 months based on this data.

Interesting thing to point out as well. Grumpy is an outlier in the opposite sense as the alliances to the far right. I think people say its hard to catch up to Grumpy in relation to vast amount of time it takes to build cities past 25.  

Being at the right of the graph means your econ program is geared towards production buildings/infra as opposed to city count per alliance. Most of the economic culture in Grumpy is production. People could catch up to grumpy if they grouped together 15-20 city 40+s. But its not enough in the game for this to happen. You all decided to group together under one umbrella and become a large production power for one sphere, which is what people must figure out how to challenge.

 

[zigbigadorlou]

So who has the best econ?

[Sweeeeet Ronny D]

29 minutes ago, Darth Tryptophan said:

Being at the right of the graph means your econ program is geared towards production buildings/infra as opposed to city count per alliance. Most of the economic culture in Grumpy is production. People could catch up to grumpy if they grouped together 15-20 city 40+s. But its not enough in the game for this to happen. You all decided to group together under one umbrella and become a large production power for one sphere, which is what people must figure out how to challenge.

We dont actually have an econ program, we have a "dont tax program", and let the members do what they want to maximize their income.  I think you are overestimating the income from production vs cash standpoint.  I personally only get about 30-40 percent of my income from resource production (depending on global radiation), the majority of my income is still stone cold cash.  I would assume smaller nations probably make a higher percentage of income from resources than we do. (tho I haven't looked at incomes of little nations in years so thats a pure assumption on my part.)

The real concern for you here should be how Rose who averages 20 cities is able to outspend everyone else in the game on city growth.  Because realistically, you arent competing with grumpy, you are competing with Rose.  Who by looking at the per nation numbers is outspending you, despite losing a war to you, and having a lower average city count.

 

[Jacob Knox]

14 minutes ago, Sweeeeet Ronny D said:

I personally only get about 30-40 percent of my income from resource production (depending on global radiation), the majority of my income is still stone cold cash.  I would assume smaller nations probably make a higher percentage of income from resources than we do.

As a C21, 76% of my monetary net income comes strictly from cash and the other 24% is resources. But I make steel with a little excess iron. Could be different for others.

 

1 hour ago, zigbigadorlou said:

So who has the best econ?

This is a question I would honestly love to see answered by someone with more economics knowledge than me lol. Or at least have their opinion on the matter (since ranking "best" based on data like this may neglect other factors and whatnot).

[zigbigadorlou]

6 hours ago, Sweeeeet Ronny D said:

the majority of my income is still stone cold cash.

A lot of the game is still scared to invest in infra and make commerce worth while. I feel like a lot of people look at the market prices and forget cash income scales with infra. But Locutus doesn't tell you to buy infra, does it?

[Thalmor]

I am downvoting this because I hate data. We need less information. We need to return to ignorance. Our decisions should be based on gut instinct alone. 

[Zei-Sakura Alsainn]

21 hours ago, Sweeeeet Ronny D said:

These numbers are out of date, grumpy has spent over 10 billion dollars the last few days picking up cities. AKA we bought like 3-4 cities the last few days.

This is because they are crappy alliances.

 

Tho to all of you that !@#$ about how its impossible to catch up to Grumpy, this gives you an idea of how long and how hard it is for Grumpy to pick up cities.  You can see out of our guys with over 40 cities, we average about 2 cities per 6 months based on this data.

It has more to do with you being as good at reading a graph as Deulos is at making coherent, intelligent speech. I'm sure you know how good Deulos is at that.

 

The graph shows how much money has been spent compared to: Total Number of Cities, Average City, and Members.

 

6 months how convenient, that's when I, the current econ, showed up. ASM was around...? 500? High 400s? I think on the onshore at that time. Had about 22 members or so, 25 including the offshore. 

Then a war happened, rebuild, and, at that point I pushed for recruitment to be opened (total membership is now 35 instead of 25, a success so far, and why the average city went down) and started an actual growth program as ASM had neither had one or a proper econ head in about 18 months.

Well, that was like... 2.5 months ago? The plan I made was supposed to start ending after 3 and it's got about, after tomorrow, 2 weeks left before it's end phase kicks in, which is with the upper tier push ending and those resources pushed downward to up their build speed. The 20s are now a 22 bar going to 25, the 25s are hitting their mark of 27, and the 30s are almost done hitting 32. 

 

I suppose I'll have go now to inform Callisto that apparently T$ and Rose are bad alliances, as she came to my DMs last week as April rolled in to congratulate me for beating the entire sphere in average member growth rate. My condolences for tainting then both with my mediocrity. 

 

As for why the money spent is so low? Almost everybody here had UP/AUP when I showed up. What's the average per member? Hard to see in my phone, it's like 600m right? 186 for c21, 517 for a 26, just over 1b for a 31. Don't forget the building of the noobs from 10 towards 20 being cheap as chips, and congratulations I've solved the case of "how can number be not big, and also not bad?" It's because there's less of us at lower city courts with huge discounts applied. 

 

Well the graph says 721 for on and offshore now and 190 built which means a starting point of 531 total. That's a 35% increase. For comparison Cata, whose always been recruiting, has enjoyed 22%. Rose has enjoyed 27%, TKR 24%, infact you'll have to tap Bourbon Street (1300 to 1800ish) with a 41% rate to comfortably sit my ass down.

 

Tl;Dr git gud, learn to read a graph maybe, and I'll keep trying to catch you pops. Looks like it can be done if I match the endurance.

 

[Sweeeeet Ronny D]

1 hour ago, Zei-Sakura Alsainn said:

"wall of excuses"

Come on now Akuryo, spend less time on excuses, and more time on buying cities!  Or dont, doesnt really matter much to me.

[hidude45454]

1 hour ago, Zei-Sakura Alsainn said:

-snip-

I was curious enough to add another tab purely for raw percentage increase of number of cities. Turns out all the alliances with the highest percentages are training alliances or incredibly low tiered alliances (shocking, I know). So I'm not exactly sure how much comparing different alliances against each other actually means, and again I suck at econ so I can't really interpret it beyond that either.

Edit: I graphed the percent increase versus the average city count per member, and this is about as clear of a trend as you can get:

TBH, my first thought was disappointment because the obvious conclusion to draw was that how good your econ is can only affect so much, but that everyone is on basically the same curve no matter what. But I don't really think that's the correct conclusion to draw either, since the graph is hugely dependent on how many newer players you have in your alliance, and while building cities at higher tiers costs a lot more, it won't make as much of a dent on the graph.

Edited April 7, 2022 by hidude45454

[zigbigadorlou]

34 minutes ago, hidude45454 said:

TBH, my first thought was disappointment because the obvious conclusion to draw was that how good your econ is can only affect so much, but that everyone is on basically the same curve no matter what. But I don't really think that's the correct conclusion to draw either, since the graph is hugely dependent on how many newer players you have in your alliance, and while building cities at higher tiers costs a lot more, it won't make as much of a dent on the graph.

You're probably not going to see the difference until you plot against city cost. The data points lie along the general trend, but there's significant deviation. Like one of those has an avg of c10 and only a 50% increase. 

Also when you consider the cost scaling, its harder and harder to differentiate at higher city counts. Basically, there is a currently unknown "ideal" curve that could be fit over the top based on cost increase and the test is to find the deviation from that curve. 

[hidude45454]

31 minutes ago, zigbigadorlou said:

You're probably not going to see the difference until you plot against city cost. The data points lie along the general trend, but there's significant deviation. Like one of those has an avg of c10 and only a 50% increase. 

Also when you consider the cost scaling, its harder and harder to differentiate at higher city counts. Basically, there is a currently unknown "ideal" curve that could be fit over the top based on cost increase and the test is to find the deviation from that curve. 

If you have any suggestions on what exactly I should be plotting/charting I'd love to hear it; I have a lot of data and not a lot of econ experience 😛

[Firwof Kromwell]

18 hours ago, Zei-Sakura Alsainn said:

Excuses to seeth on Deulie & Ronny

Woah there bud, are you sure they are the ones whom cant read cuz I swear if scroll up I saw Deul & Ron having a constructive & concise back n forth w/ some others chiming in respectfully. In that case, that would hint that anyone whom associated w/ the convo stream above cant read except those you agree w/ which im not surprised comin from a ego-bust like yourself. Also if what you said was true that yall had a growth plan in place, then the data would have shown in the graphs due to the timeline n "data" you given but thus isnt the case. ASM is still shown in almost all tiers (& quite more so exempt in the low-med tiers), more stagnation. Hell even your aa page graph shows it wasnt until a couple weeks ago yall started on an actual consistent higher sloped uptick in growth, slow n steady pace outta da top 50 to a more notable rise into the top 40 in the mid 30's. 

IDK thou, maybe I cant math, especially graph reading even w/ taking Advanced Algebra & AP stats (w/ acing most assignments in both, aside math being a strong suit of mine). One thou even outside of taking such classes should be able to be very easily to observe such thing w/ ease (like me who just showed my younger sister, whom doesnt play nor want to know much of pnw/what I do, these graphs & agreeing w/ the basis of my reply).  Then again, you as well a few of yo buds jn ASM have in the past over amped yourselves while statistics show elsewise. I would say something else about yo homies regarding intake of facts n the truth but it be too ooc althou I do hope ppl uptake on the hints lying around.

[zigbigadorlou]

23 hours ago, hidude45454 said:

If you have any suggestions on what exactly I should be plotting/charting I'd love to hear it; I have a lot of data and not a lot of econ experience 😛

Your plot looked a lot like a y=k/x plot so I just plotted cities vs the inverse of % growth. Taking out oblivion who is a huge outlier (1/ 2% growth makes a really big number). I also took out grumpy as less of an outlier.

To me, that's a bit easier to interpret. The average is therefore % increase  = 17 / (the average city count) - 0.64. So as your average city count goes up, the % increase (the rate of increase) goes down. This implies 17% growth per city. However a lot are far below this line. 

So anyone above the line is better than average and anyone under is worse than avg. The outlier is Otaku Shougaku. Note that the right end of the spectrum are all the lowest averages.

Here it is after cutting out anything under average city 14 (with Grumpy). Very different slope at higher city counts, but probably more applicable. So there's a growth of more like 5% per city after city 14. 

[Sweeeeet Ronny D]

The one variable that needs to be taken into account is war activity.  If someone got rocked in a global war, I would expect to see your nation growth dip, since you would need to focus on rebuilding, and warchest restocking before you go back to city growth.  But if you are below that line, and have been war free for the last 6 months, or got off really easy in a war, then yeah, your alliance has been doing a poor job prioritizing nation growth.  Does that mean your econ is garbage? not necessarily, but it does point in that direction.

[hidude45454]

On 4/8/2022 at 6:11 AM, zigbigadorlou said:

Your plot looked a lot like a y=k/x plot so I just plotted cities vs the inverse of % growth. Taking out oblivion who is a huge outlier (1/ 2% growth makes a really big number). I also took out grumpy as less of an outlier.

To me, that's a bit easier to interpret. The average is therefore % increase  = 17 / (the average city count) - 0.64. So as your average city count goes up, the % increase (the rate of increase) goes down. This implies 17% growth per city. However a lot are far below this line. 

So anyone above the line is better than average and anyone under is worse than avg. The outlier is Otaku Shougaku. Note that the right end of the spectrum are all the lowest averages.

Here it is after cutting out anything under average city 14 (with Grumpy). Very different slope at higher city counts, but probably more applicable. So there's a growth of more like 5% per city after city 14. 

I like this a lot, added it to my sheet so people could mouse over the dots and see who's who

And yeah Oblivion was a huge outlier when I was first compiling the sheet since we were kind of intentionally holding off building; we finally started building cities again in the middle of it so at least the sheet did something lol.

[hidude45454]

Part 2 as requested, with about double the alliance count and additional sheet comparing gov with non-gov: https://docs.google.com/spreadsheets/d/1VOPn7LhHvOOfK41vU_fCzZMM-sYmnCiTFLjPmJgl7uQ/edit#gid=165451901