Stats 'n' Facts

[DeletedUser]

I have been a little depressed recently that some players seem to not think much of my contributions to the World 73 forums. This is partially why I started this thread up. Before I became more known as a troll and provocateur, I actually posted primarily analytical and statistical material which was my main passion before I realised it was easier and more rewarding to annoy people. Although it has been a long time since I have actually posted something in this vain, I shall give it a try. Some of you oldies might know that the name "Stat's 'n' Facts" originally came from a huge number of world statistical analysis threads conducted by Dark Chaos for World 1 and 2. As I am quite uncreative, I decided to transplant it to World 73 for the purpose of starting this thread. This will be long, this will be boring to most and perhaps interesting to others. It will contain no opinion, only statistics based off data either collected directly from the game or from the semi-official statistics site www.twstats.com. It will almost certainly contain large amounts of mathematical workings. It may also contain graphs. Those graphs may also be scrappily put together by a person who has deleted all graphing programs other than Excel. It will also probably have little structure, I will purely be starting from the above two sites and will follow the rabbit hole to see where it leads.

With all that rubbish finished, lets get underway. I shall be editing this on the fly as I make more stuff for it rather than making a completed product as I want to be able to incorporate stuff the public might be interested in or answer any questions as I go along.

Stats 'n' Facts

When analysing a world, one should always start with the tribes. This is Tribal Wars afterall.

A few questions.

What are the distribution of tribes within world? What is the average size of a tribe as of today? What makes the top tribes tick? What percentage of barbarian villages do top tribes take and how does that compare to their counterparts? What about internals? What about their player base? Are they very top heavy or relatively evenly distributed in terms of size compared to other top tribes?

Well better get answering, I will probably answer a few I haven't put here too if I think of them on the journey.

Part 1 : Looking at tribes on a global scale

So, lets start by looking at tribes from a global rather than on an individual basis. Firstly I will have a bit of a look into the size of various tribes. I decided the simplest way would to divide the size into six basic brackets in terms of groups of two hundred points up to the three biggest tribes which are larger than 1 million points. Made this into a chart:

Table 1: Number of tribes at Different size brackets

There seems to be a slight problem. It would make for a stupid looking graph with most of the points being non-existent. So I did a bit of tweaking. Firstly I changed the criteria so that my potential graph would show the number of tribes over certain point thresholds. This meant 3 tribes over 1 million, 6 over 800k and so on... I also added some points between 0-200.000 points to flesh out that part of the graph and made the following:

Graph 1:

[SPOIL]

As requested, the same graph in a logarithmic scale. It needed even more to make any sense so I added 5 more in (100, 500, 1000, 5000, 10000 pts). Just a notice, although it does give an exponential trend however this discounts the total number of tribes (as it quite rapidly flatlines towards 1015), which is why I chose not to use it originally.[/SPOIL]



The numbers next to the points are the specific numbers of tribes larger than the number of points on the x-axis. Now if I was really clever, I would be able to have a good idea of the mean size of a tribe in world 73 by calculating where the slope of the derivative of a trend line equation is -1, however I have forgotten how to math and could probably do it if I had an hour to refresh but moving on...

Now we have got a good idea of how the tribes are distributed as a whole, lets narrow it down a bit to just looking at the top 100 (eventually will go down to top 10 and then look at player statistics). At this point, I am going to post this and edit it and will be working on it over the next few hours.

Oooh, I see the images are blurry. Sorry, too bad.

Anyways, onto part 2.

Part 2: Looking at the top 100 tribes

When looking at the best 100 tribes in the world, where better to start than looking at the members. On a very basic level, this is questioning trends such as how many members are in top tribes. I had a bit of time so why not make a graph of that:

I'm not sure I am happy with how this graph came out, but I think I will use it anyways. Each different colour of grey represents a different member count with 0-10, 11-20, 21-30 and 31-40 respectively. It doesn't factor in the various family and academy tribes. A few things that are interesting to notice, there is only one tribe that can be classified as 0-10 members, the rank 55 tribe DbD. Aside from that, 8% of tribes in the top 100 contain 11-20 members, 21% contain between 21 and 30 with the vast majority being between 31 and 40. The lower ranked tribes seem to be a bit more elitist on member count or perhaps not fully fleshed out yet.

I will probably come back to this later but for now I will skip over to top 10 tribe analysis.

Part 3: Looking at the top 10 tribes

This is where it starts to get interesting. Why top 10 instead of top 20? I am lazy and 10 is a smaller number than 20. It's also easier to make graphs and charts out of. Anyways, lets try and chart some basic information for each tribe and put it in a nice and neat table.

Here is a basic table for starters:

Ever wondered about the distribution of member size within tribes?

Just a note on this one, by top 100 player I mean rank 21-100, top 500 = 101-500 and so on.

What about a table to show who is throwing nobles at what (please note I have not included family tribes due to not knowing the tribes well enough):

 

[DeletedUser]

Reserved.

 

[DeletedUser108821]

Can you put the graph on a logarithmic scale? If I'm right it'll show a clearer trend.

 

[DeletedUser]

Reserved.

+1 already

Edit: Or not

You must spread some Reputation around before giving it to the nemesis123 again.

 

[DeletedUser]

This looks fun, I'll be sure to invest some time into looking at your stats here nemesis. I hope you keep it up a bit longer than that world73 blog did.

 

[DeletedUser]

Can you put the graph on a logarithmic scale? If I'm right it'll show a clearer trend.

Yes, I did consider using a logarithmic scale. Have put it in a spoiler under a graph as well as a bit of an explanation of the trend and why I didn't use it in the first place.

 

[ampac]

Nemesis u nub i dont understand what u posted again......................maybe im drunk or maybe i got a short attention span

 

[ashoka1]

Nemesis u nub i dont understand what u posted again......................maybe im drunk or maybe i got a short attention span

+1

 

[DeletedUser]

Just a note, the most recent graph is slightly inaccurate due to ODA being 30 members instead of 31 which I had them down for for some reason. Will probably fix it in the coming hours but doesn't affect trend too much.

 

[ALessonInPointWhoring]

I skew Randoms stats rather negatively as well for the moment, me being 140 points and ranked outside the top 10,000 really gives a beating to average player size.

 

[Deleted User - 4669627]

I have been a little depressed recently that some players seem to not think much of my contributions to the World 73 forums. This is partially why I started this thread up...

ty for contributing, especially this bit considering our nobling practices have been called into question recently while we even now continue to lead this particular twstat

From the last 7 days

Rank	Tribe	Conquered Villages
1	CODEX	68
2	Random	65
3	IKEA	65
4	Joker	55
5	WET	52
6	FIRM	50
7	ODA	45
8	Mant	43
9	None	38

admittedly we have a lot of barb conquers, compared to random for example
random began day 1, we began day 10 or 11, i forget
many of the players near us still have only 1 village and quit when they get cleared, but even we are moving out of that stage now
idk what is the issue with wet and firm, perhaps they have already moved into the accounts going inactive stage? but firm's internal % is extremely low

our internal rate was also called into question, and the fact that majestic knight had just joined us, then quit, was used to try to pervert the stats
admittedly though, we have a lot of players mergign accounts to co-play, virtually every account we started with were single players (cuz of the restrictive nature of our recruitment criteria before the world started, basically, you had to be my personal mate in a previous world during a war), we account sit each other till we have about 3 villages each, then began merging and co-playing

2014-03-17	CODEX	5.	+39	+1,014,844	+225
2014-03-11	CODEX	5.	=37	+665,292	+164

in these 7 days we have added 2 members and 61 villages, the 2 members have 7 villages between them
our total point gain? 349,552
growth %? 34.45?

2014-03-17	Random	1.	-30	+1,151,178	+221
2014-03-11	Random	2.	+28	+796,750	+160

random added 2 players (same as codex) with 10 villages (3 more than codex) between them gaining 61 total (same as codex)
total point gain? 354,428 (basically the same, obviously them being closer to the core means larger villages)
growth %? 30.18?

i have not done the other tribes, but if they also wish to compare stats, we can
maybe nemesis would like to double check my stats?

also i see in some threads random still is talking about our ec merge, despite their own more questionable recruitment
i have not counted the number of players random has invited, nor have i counted my own
we started with less accounts and held off ingame recruiting for a longer time
when we did open recruitment, we recruited as much as random had already done, except we did it in 1 shot

that might be an interesting stat nemesis

 

[sp.ag]

You and the tribe may of started XX amount of days late but how many of the memberbase started at the same time as Random members? or near enough a day apart so not sure that stat can be used to compare the tribes.

 

[DeletedUser]

random added 2 players (same as codex) with 10 villages (3 more than codex) between them gaining 61 total (same as codex)
total point gain? 354,428 (basically the same, obviously them being closer to the core means larger villages)
growth %? 30.18?

Growth percentage is a statistic used when you want to give the benefit to the smaller tribe, so good job for having a higher % growth than someone that started larger than you. If it were the other way around, however, then you'd want to be worried.

If you actually had better stats, you could post JUST the numbers, without your commentary, and they would stand alone and tell the same story.

i have not counted the number of players random has invited, nor have i counted my own

convenient...

we started with less accounts and held off ingame recruiting for a longer time

And you end up with more accounts, after merging accounts between players that knew they probably wouldn't continue playing the world on the account they started with, and that they'd merge with someone later...

Joining a world as individuals with a plan to end up coplaying with one another is very questionable. Not illegal per se...

when we did open recruitment, we recruited as much as random had already done, except we did it in 1 shot

Yet you're just shy of the member limit... and random? Comfortable 10 positions available - leaving room for when we expand and pick up our sleeper agents.


We know you have an ego the size of a planet, but boasting about mediocre stats doesn't make them impressive, and it doesn't make you any less insignificant in the long run.