NESing by the Numbers

Well, leaving out the zeros, the median is 4.

Also, interestingly enough, only 2003,2004,2007, 2010, 2011 show statistical significant in the number of updates per NES, all others fall within the standard error, with 2003,2004, and 2007 being spikes, and 2010 and 2011 being lows. As far as number of updates per NES, one cant say there was a decline until 2010.


Except for in 2003, the number of NESs per year fall within 2 standard deviations (STD = 30.7) of the average (99.4) so its hard to say that there is a statistically significant decline in the number of NESs

Though, I'm sure there is a better probability distribution than normal to fit updates per NES, but I don't remember much from that part of statistics. OTOH, A normal distribution fits NESs per year nicely.

NES/Year Average: 99.4
STD NES/Year: 30.6
Average Updates/NES: 4.96
STD Updates/NES: 8.986

2002
Number of NES: 88
Standard Deviations from Average: -.372
Updates: 465
Updates/NES 5.28
Standard Error: .958 Updates/NES

2003
Number of NES: 176
Standard Deviations from Average: 2.5
Updates: 1099
Updates/NES 6.25
Standard Error: .677 Updates/NES

2004
Number of NES: 115
Standard Deviations from Average: .509
Updates: 690
Updates/NES 6
Standard Error: .838 Updates/NES

2005
Number of NES: 112
Standard Deviations from Average: .410
Updates: 472
Updates/NES 4.21
Standard Error: .849 Updates/NES

2006
Number of NES: 105
Standard Deviations from Average: .183
Updates: 457
Updates/NES 4.35
Standard Error: .870 Updates/NES

2007
Number of NES: 72
Standard Deviations from Average: -.893
Updates: 466
Updates/NES 6.57
Standard Error: 1.06 Updates/NES

2008
Number of NES: 104
Standard Deviations from Average: .15
Updates: 483
Updates/NES 4.64
Standard Error: .881 Updates/NES

2009
Number of NES: 83
Standard Deviations from Average: .53
Updates: 341
Updates/NES 4.1
Standard Error: .986 Updates/NES

2010
Number of NES: 78
Standard Deviations from Average: .698
Updates: 249
Updates/NES 3.19
Standard Error: 1.02 Updates/NES

2011
Number of NES: 229
Standard Deviations from Average: 1.25
Updates: 229
Updates/NES 3.75
Standard Error: 1.15 Updates/NES

While the numbers indicating a "decline" in NESes and NESing are nice and all, they fail to take into consideration the NESes themselves. Nowadays, NESes are far more complex, or at the very least, quality-controlled. Some even take months to develop, such as SysNES2 or DachNESIII for example. It is very rare instance now where someone will just start a NES on a whim (Sheep comes to mind in particular here) Sure, there may be a decline in the total quantity of NESes, but that doesn't mean the total quality and effort put into NESing has declined. I myself have scrapped several ideas for a second NES simply because I was unsatisfied with the early results. (or I have very high standards )

For a practical use, it'd be interesting to see if a pre-NES thread has any correlation to the success of the NES.

erez87 said:

Also Mods should be counted by number of updates not by number of posts. That's just show how spammy they are.

Click to expand...

Not all updates are equivalent. This isn't a dig at minimalist updates, they're very good at what they set out to do, but something like Perfectionist's updates are worth >4 of many other mods.

Also, the "decline" seems to be in the number of new NESes, a very different stat from the NESes being played.

Does anyone know how to analyze a logarithmic distribution? After generating a histogram, this appears to be the best distribution for updates per nes but I have no idea how to analyze one.

Well, I'll state once again, my interpretation of the numbers is that we've reduced the number of dead on arrival NESes. Overall, NESes have increased in quality. Another number which we could theoretically use to track activity would be the number of players that have been active in each year/month. This be may difficult to determine however, as it would take substantial reading of all the threads once again. If we knew the respective number of players (and checking the census threads that have occurred on occasion) we could probably gauge more certainly whether we are truly declining or simply just evolving.

This totally blows my mind and outscaled anything I had hoped to achieve with my qualitative Census. . I am doing AP stats right now, and I'd love to dig into that sweet, sweet data during spring break, or better, summer vacation.

LINESII Part II

Impressive dataset!

Boundless said:

Never properly connecting with the Civ4 community could have been the start of the decline.

Click to expand...

We did get to FFH'ers from there. Ekolite, Immac, Seon?, Thomas(I think?) are few examples.

I came from the Civ 4 community

Yeah, I don't think we failed to connect with Civ4, and I think overall switching to our current location probably has helped bring in new blood.

What is the Median including 0 updates?

And what is the Median per year?

Northen Wolf said:

We did get to FFH'ers from there. Ekolite, Immac, Seon?, Thomas(I think?) are few examples.

Click to expand...

I actually think most of us who could be considered FFH'ers were actually brought here from Forum games by Abaddon.

North King said:

Is there some quick way to check the wordcount of a thread or post? Otherwise that's going to take forever.

Click to expand...

Write a script in a good string manipulation language like perl or python, then do it in seconds if you have a nice fast connection, you could even do words per user id.

I'll write the script when I get home - is there a particular thread someone wants word-counted first?

I reject any statistical tabulation that puts Sheep in tier 3

Edit: didn't see the table included links to the threads, I'll just do all of them then.

erez87 said:

What is the Median including 0 updates?

Click to expand...

2

5char

There hasn't been a steep decline from 2003. 2008 shows a huge rise in NES numbers, and 2009 and 2010 are still superior in numbers to 2007.
Trying to fit a linear regression on these is pretty useless as what we see is a growth, decline, then smaller growth and decline again. Fitting a polynom could be better but you'd need at least 4th degree and that wouldn't tell us much anyway.
It's easy to find out that the phenomenon starts low, grows and then people leave.

Interesting datapoints are:
2003: Peak in number of NESes. 2008: Smaller peak. What happened in 2008?
2006: Peak of number of posts for this year's NES. Smaller peak in 2008. This is quite interesting, as 2006 NESes are less numerous than those of 2003. There are twice as many posts as in 2003 for a lower number of updates.
Updates per NES peak in 2007, when the number of NESes was lowest.
The number of posts per NES also peaks in 2007:
2002 226
2003 216
2004 357
2005 443
2006 642
2007 663
2008 494
2009 411
2010 336
2011 290

From this I would say there are 3 turning points:
2003 sees the maximum volume in terms of started NES.
2007 is lowest in number of NES, but maximum in number of updates and number of posts per NES.
2008 sees a lot of new NES come by, but from there the number of new NESes diminishes. The number of updates per NES and posts per NES also shrinks.
I wonder how much of the posts/NES diminution is due to groups, where a lot of traffic has been happening and which didn't exist in 2003. I'm not clear about PreNES threads which sometimes turn into Out Of Character threads.
Note that 2008 saw TheJopa's FfHNES, which drew a few people in here, like me. It certainly drew some people from Civ IV at that time.

Well, I'm prepared to present my theory on this but, I'd like to start by agreeing with Disenfranchised. Sheep should not belong on any tiered system, and if there's one thing I've learned through this whole process, it's that I don't dislike any of you. In fact, I no longer feel hate at all, except towards Sheep. I do in fact hope that by the sheer force of hatred, he will never return. Speaking of Sheep, it's his departure that likely caused the decline in numbers of NESes, as there has been no one like him in terms of creating miserable rip-off failures of NESes. Sure, I've made a lot of NESes over the years, and most haven't made it very far. Sheep on the other hand, in a single day created and killed three NESes. Then we add in the blatant plagarism and copy/paste techniques of his moderating style. Let's take a look at his statistics, which I'd argue are worse than any except the occasional one-off moderators who appear every now and then.

Sheep
Starting Date of first NES: 11/15/2002
Ending Date of last NES: 6/19/2009
Total Number of NESes: 47
Total Number of Posts in All NESes' Main Threads: 7,801
Total Number of Updates in All NESes: 68
Average Number of Posts per Main NES Thread: 165.98
Average Number of Updates per NES: 1.45

For my first major point for all NESers, if Sheep returns again, avoid his NESes like the plague.

The 2003 spike

Speaking as someone who was there, and fairly recently reread all of these NESes, 2003 was a hotspot of activity for two primary reasons. First was the success of Return to Our Roots NES 2. This was the first epic NES which reached 5,000 posts and spawned an incredible number of imitators during the whole of its existence, which was primarily in 2003. It also created the "Fresh Start" genre, and then was followed by the equally successful stJNES 2 and stJNES 4 which also took place in 2003-2004. There were many, many imitators of both NESes, some fairly successful in their own right. The second main reason was the continued height of the board game NESes, which featured many updates, fantastic moderators such as uknemesis and RoddyVR. These NESes also lasted significantly less time than NESes do today, allowing time for the creation of more and more. The next reason is Sheep. He created 21 NESes over the course of 2003 alone. Take from taht what you want. Lastly, I'll point out that NESes had a different standard of life expectancy. If an NES didn't update for the course of a week, it was considered "dead" by all involved. An NES that only updated once a week would be considered mind-bogglingly slow by all the players. Therefore NESes burned bright, but died faster, allowing for the creation of more and more.

The 2007 Dip

The dip in 2007 could be explained for a number of reasons, including a population change among the playerbase. I've been told that a large number of the old player base graduated from high school in 2007, and went on away from NESing after that. Not sure how accurate that statement is though. Also, there were a number of NESes in 2007 that were continued from 2006, including LuckNES 2, DNES 2, among others, including a few of my own first successful NESes since RTOR 2.

The "decline" of NESing

I truly don't think we're declining. I think that with the constant support of the New NES Idea Thread and older players, we're simply refining our methods and reducing the amount of NESes created on a whim. If someone chose, we could figure out the number of players in different years, but that would take a lot of effort. The only true way to gauge if we're declining is if we're bleeding players.

These are my own stats, taken from EQ's data as they appear in the first post.

Milarqui
Starting Date of first NES: 20/2/2009
Ending Date of last NES: Still running
Total Number of NESes: 7 (MilarNES: Birth of Civilization, MilarNES II: Civilizations of the World, MilarNES III: Raw & Stock, MilarNES 3: Battle for Wegener, The World is Yours, Evil Genius NES, Viva Trópico 2: A Pure Politics NES!)
Total Number of Posts in All NESes' Main Threads: 1535
Total Number of Updates in All NESes: 34
Average Number of Posts per Main NES Thread: 219.286
Average Number of Updates per NES: 4.857

merciary said:

I actually think most of us who could be considered FFH'ers were actually brought here from Forum games by Abaddon.

Click to expand...

Proof he did some good around here

Less Neses, posts and updates does not mean we have become more "quality" oriented. It means there are less Neses, less updates and probably less Nesers.

I wonder if the Average of updates and the Median of updates have changed through the times. If it increased than I am wrong, if it didn't...

Number of updates does not make a NES higher quality in of itself, if that's what you're implying.