Mag Talk Weekly Shonen Jump [2012] - Discussion and TOC Talk

[Newkerzy]

True. True. But IMO, I'd like early chapters. Of course, just my own personal opinion.

 

[Schabrak]

True. True. But IMO, I'd like early chapters. Of course, just my own personal opinion.

It'smalways an "early" chapter as mentioned before. Having to wait 1 1/2 weeks for the next one sound far more agonizing than the chance to get it two more days earlier.

 

[Kaiten]

But that's what statistics is about too: finding a "pattern" in things with apparent randomness, like opinions, throws of dice, or, in this case, ranking positions.
Of course that's not a proof of a sucess/failure, but it's an interesting information.

That's not how statistics work, smart guy

Statistics are a mathematical model of numerical trends. Statistics are not used to find a "pattern in randomness", randomness is a pattern. And we are not discussing statistics, we are discussing probability, the percent chance of a specific outcome occurring. The discussion is whether an outcome, long term success of a manga, can be determined by a new manga's first ranking only. To do so disregards the ranking of all subsequent chapters, and the effect they have on future success. If K-dom's analysis was repeated using the average rank of the first eight chapters of a manga, rather than only the first, the probability of success or failure would be greatly altered. Common sense dictates that determining success based on multiple rankings would be far more accurate than the same study based on only one ranking.

Sorry K-dom, a few issues with your findings:

• How are your percentages effected by increasing the number of ranked chapters studied? What happens If you look at the average rank of the first 2, 5, or 8 chapters?
• What happens if you reduce chapters achieved to a number below 100, what if you increase to a number higher? What does this say about the percentage of series? Could the number of chapters achieved, rather than rank, be a determining factor in percentage of series that survive?
• How many series debuted between issue #1, 2005 and the issue you ended the study with? What percentage lasted 100 chapters or more? How does the total population (all series that lasted 100 chapters or more) compare to the sample (how many series lasted 100 chapters or more based on ranking of chapter 1)? How do they compare if the number of ranked chapters included in the survey is increased?
• Minus series ranked 7-10, your data follows a perfect bell curve. Why is 7-10 an anomaly? How is this effected by no manga initially ranking seventh?
• How is the math effected by the vast majority of series debuting between 6 - 20? Shouldn't 2-5 and 21-22 be eliminated as outliers? We know there is a high probability that a series is likely to survive if it debuts in the top five. We also know the probability of survival is extremely low if a series debuts in the bottom 2.

I redid the statistics, accounting for outliers, and the results are much more informative:

Rank	Nb serie	Nb > 100
2	1		1
4	3		3
5	2		1
6	7		4
7	0		0
8	6		1
9	3		1
10	6		1
11	3		1
12	5		2
13	8		3
14	4		2
15	6		0
16	5		1
17	5		0
18	7		1
19	2		0
20	4		0
21	1		0
22	1		0
CC	9		4

• Total Number of Series: 88
• Total Number of Series With >100 Chapters: 26
• Percentage of Series Lasting >100 Chapters (26/88)= 29.5%
• Rank 2-5 (1+0+3+1/1+0+3+2)(outlier)= 83%
• Rank 6-10 (4+0+1+1+1/7+0+6+3+6)= 32%
• Rank 11-15 (1+2+3+2+0/3+5+8+4+6)= 31%
• Rank 6-15= 31%
• Rank 16-20 (1+0+1+0+0/5+5+7+2+4)= .09%
• Rank 21-22 (0+0/1+1)(outlier)= 0%
• Color Pages (4/9)= 44%

Determining a population mean and removing the outliers clearly has a huge effect, normalizing the results:

29.5% of the manga serialized since 2005 have >100 chapters. 83% of manga that debuted in the top 5 exceeded 100 chapters, as expected an outlier much higher than the population mean. Only .09% of manga that debuted between 16-20 survived, also to be expected. No manga serialized in the bottom to reached 100 chapters, an expected outlier. 32% of manga that debuted between 6-10 reached 100 chapters, 31% of manga that debuted between 11-15 survived. Combined, 31% of series that debuted between 5-15 reached 100 chapters. That establishes an expected linear regression, with the odds of reaching chapter 100 shrinking the lower the initial rank. Survival rate for the majority of series (31%) reflects survival rate for the population as a whole (29.5), accounting for a reasonable margin of error. Survival rate for very high ranking, and very low ranking series are clearly demonstrated as outliers, higher and lower than the population mean.

Obviously these numbers will be heavily effected by adding additional rankings per title. I do not have the time or interest to undertake this, and the results should be obvious. Maintaining a ranking between 6-15 should increase chances of reaching 100 chapters, rising into the top five should increase the chances exponentially, falling to 16 or below should decrease odds.

line of regression:

y= -.17X+3.5 (X Axis=Ranking, Y Axis=Nb series w/ >100 Chapters)
y= .45X - 1 (X Axis=Nb series, Y Axis=Nb series w/ >100 Chapters)

For anyone wondering: yes, statistical analysis is something I do in real life, generally for things more important than mangas.

 

[Zeromcd]

It seems like something very big will be announced next Saturday. It will also be announced in Issue #33.

http://shonenjump.com/j/countdown/

 

[xi0]

It seems like something very big will be announced next Saturday. It will also be announced in Issue #33.

http://shonenjump.com/j/countdown/

It seems to be emphasizing "D"? No clue on that front

 

[Zeromcd]

It seems to be emphasizing "D"? No clue on that front

It says at the bottom when you think of "D" you think of One Piece or Dragon Ball.

 

[FrostyMouse]

Who knows. The D message confuses me even more, unless it'll be a crossover OP/DBZ series/movie/something.

 

[B'arzz]

It seems like something very big will be announced next Saturday. It will also be announced in Issue #33.

http://shonenjump.com/j/countdown/

Thanks!!! i wonder what could be maybe one piece? or something of the old school

 

[Niko_kun]

It seems like something very big will be announced next Saturday. It will also be announced in Issue #33.

http://shonenjump.com/j/countdown/

maybe something about D.Gray-man? *-* really hope so ))

 

[FrostyMouse]

maybe something about D.Gray-man? *-* really hope so ))

That could be something positive, perhaps the end of DGM, although DGM's a SQ series now, so news about it probably wouldn't be in Jump, but you never know.

 

[Estranho]

Would be really good if it was a new Jump Stars DS game, but I couldn't think of any relation with the D letter in this case.

 

[Kaiten]

maybe something about D.Gray-man? *-* really hope so ))

Wrong magazine

It says at the bottom when you think of "D" you think of One Piece or Dragon Ball.

Only thing I can think of is news about the upcoming One Piece movie. That seems like a stretch though.

 

[Galactic Tomahawk]

Could be D.Gray-Man coming back to Jump but that would be the most boring thing ever to make a countdown for.

 

[Estranho]

It says at the bottom when you think of "D" you think of One Piece or Dragon Ball.

The message on the bottom is just a feed for tweets with the hashtag. I just tested myself.
So, nothing oficial coming from there.

 

[Niko_kun]

That could be something positive, perhaps the end of DGM, although DGM's a SQ series now, so news about it probably wouldn't be in Jump, but you never know.

Yeah, i thought it would come back to jump

Wrong magazine

I know dgm is from SQ but who knows, they're both of shueisha and dgm was from jump once

Could be D.Gray-Man coming back to Jump but that would be the most boring thing ever to make a countdown for.

Yeah, would be but its not a BIG announcement for a countdown right??

 

[xi0]

I think the safest guess is that it's some sort of special issue of Jump or the like.

 

[Zeromcd]

I was just reading through some comments about the D on 2ch and I thought this one was really good.

Digital Distribution

 

[TemplateR]

Digital Distribution

Yeah, that was aslo my thougt. Perhaps a worldwide Digital Distribution in english-version ?^^

 

[Newkerzy]

Don't count on it.

 

[Nii]

Don't count on it.

*sigh* Could you say more about it than 4 words, f.ex. why you think so?

I doubt a worldwide or even a only japanese Digital Distribution. In the end it's all about money and I'm pretty sure Sueisha/Jump would fear selling less magazines if you could just read everything on the net legally and easily. Even if you need to pay for that feature someone would find a way to copy and paste it - sounds like a very risky move from Jump's side and I haven't had the impression that they are really that ambitious.