Idol Thoughts: Is it possible to have an “accurate” Request Hour?

I will probably write more about AKB48’s Request Hour after the top 200 songs are performed later this week, but this year deserves some special notice because, for the first time, they are doing a request hour 1035. This is every single AKB48, SKE48, NMB48, HKT48, JKT48, SNH48, SDN48, solo and unit songs, ranked by fan votes. Since performing all of these songs live is pretty impossible, everything up to 200 is put on YouTube. This is a fascinating look at AKB48 for anyone, so I’d consider taking a look.

The first thing that’s pretty easy to notice is how much the 48 groups have changed in terms of structure. The low ranked songs from 2006 – 2009 are almost all stage songs. The low ranked songs from 2010 – 2014 are usually the B-sides to singles. Any AKB fan could tell you that the focus has moved from the theater to other venues, but this is a clear reminder that stage performances have taken a backseat.

However, the main thing I was complaining about is that this ranking does not seem to be about quality at all. I know that must seem incredibly obvious, since this is a fan ranking, but it got me wondering if there was a way to “fix” this.

Essentially, voting for Request Hour is a second Senbatsu Sousenkyo – it’s a popularity contest for members, hidden in the trappings of voting for songs. People tend to vote for the songs that feature their favorite member most heavily, which is why Itoshiki no Natasha ranked so highly for quite a while (because Sashihara Rino fans were voting it up high). There are always some very highly ranked stage songs because of that very reason.

However, is there a way to have a more accurate fan vote focusing more on quality of songs? Fan voting is always going to have an element of bias towards favorite members. I’ve listened to many more Watarirouka Hashiritai songs than I have listened to no3b songs, for example, because I much prefer Watarirouka as a group, and that’s the case. Newer songs are going to have an advantage because current fans will have more likely listened to them than older songs. Older fans are smaller in number, and many of them may have moved on from 48 groups.

Further, there is an incentive to vote for recent songs because of the changing lineup of members in the groups. Say your all-time favorite member graduated. You might vote for the song featuring your current member, because you know that at least they will be there to perform the song.

I don’t think there’s a way to get away from this affecting the rankings, especially the top spots, as long as you have fan voting. However, ranking every song gives an interesting opportunity to see all the songs voted higher. I’m not sure if AKB will do this again next year, since they are making a big deal that this is the 10th anniversary of AKB48. However, if they do it again, the “fix” that might make things more interesting – allowing each vote to vote for multiple songs.

Now, I doubt that AKB48 will do this. Since votes are limited to fans who either members of a subscription or for fans who bought a copy of the most recent single, there are financial stakes in the request hour. That said, there are financial stakes for the fans, too. If you are going to buy a single and have a limited number of votes based off of what you are willing to pay, you aren’t going to throw away those votes

However, if you got a ballot that allowed for, say, 10 songs without repeats, you might be inclined to vote for songs you like to listen to in addition to songs that feature your favorite. Weighting these might make the results even more reliable; have fans rank the songs 1 – 10, and weight it at 10 points for 1, 9 points for 2, etc. I know this is how a lot of foreign fans tend to weigh voting systems, but it works pretty well. So that way, fans could definitely vote for their favorite member’s biggest song, but they might be inclined to give their #10 vote for a song they just like to listen to.

The biggest issue with this list is the lowest songs have too many ties; I’m assuming they’re songs that got 1 vote, then two votes. This system could help break those up and get a more solid ranking, as well as allowing for more popular group songs get further (Party ga Hajimaru yo and Wasshoi B do not belong in the 400s, I’m sorry).

No system is going to be perfect for getting an accurate ranking of 48 groups songs, especially if you want a ranking that is about quality rather than who is in the song. However, allowing for more songs per voting system might give people the perceived ability to vote for more songs, and get a more varied and more interesting Request Hour.

1 thought on “Idol Thoughts: Is it possible to have an “accurate” Request Hour?

  1. Pingback: Recommended Jpop♀ Reading: January 21, 2015 | Idolminded

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.