Aniblog Tourney Thoughts

Since I have basically no chance of making it past the third round and have made it as far I wished, I’ve decided to post my thoughts on (or maybe I should say problems with) the aniblog tourney rather than advertise. I know that these types of posts have been everywhere, so I’ll try to make it worth your time. Blah blah circlejerk or ramblings about vote manipulation or vote begging…none of that here.

Personally, I believe that asking a friend for a vote is perfectly fine because the tourney has no restrictions on participants and the rules say something to the effect of “vote for the blog you’d rather read.” I’m fairly certain my buddy would rather read my blog than my opponent…simple logic. That aside, let’s look at things a bit more analytically…keep things objective. Sorry…this is gonna be long.

My first qualm with the aniblog tourney is pretty basic, but it is something that doesn’t seem to have disappeared with the new voting system…a relic of my AP Statistics class in high school. Polls should always have a random ordering for the choices. Given two choices (and in this round’s case, four), it’s much more likely for someone to pick the first one given no strong feelings for any other choice. It’s not really a big problem to implement this sort of thing, but it does a lot. Don’t believe me? I ran the numbers. In the first round, 62% of the blogs who won were the ones listed first.

Yes, I understand that the posts have the headers for each matchup that say “XXXX vs. YYYY”, but that isn’t an excuse…it just sounds like laziness to me. Does that even have to be there? Even if it does, I think it still might be helpful to have them below the poll rather than above it so that the first thing a voter sees isn’t that ordering.

A view of my own stats page. The weekends are the darkened columns.

Next up on the docket is something that I actually pointed out to someone before the tourney started, getting the response of something to the effect of “oh, it won’t be a big deal…” Time frames. Each match appears for two days, with new matches going up every day. People have stuff to do, and internet traffic reflects that. Typically, more people are online over the weekends than the middle of week (Wednesday/Thursday), which skews samples.

Even if you ignore that, there comes the issue of motivation. Four matches go up on one day, or in this round’s case, one match of four blogs. It’s just not possible to research every blog in every matchup with new ones coming each day. And even if someone had the time, where’s the motivation? Why not just wait for the next set of blogs? With how long the tourney has dragged so far, it hardly seems logical to use an argument of keeping interest in the event.

Let’s take a step back and talk about some smaller things (and by that, I mean things that I don’t feel like talking about for a full section individually). First up, participant selection. The organizers created some rules for qualification (no photo blogs, no translation blogs, etc.) and then proceeded to ignore them. Nothing more to say on that.

Next, settling ties. Rather than do something like create a tiebreaker round, the organizers choose to allow both blogs to advance. I think that’s just unfair to punish the next round blog, which actually had nothing to do with creating the tie. For a similar reason, making the tied blogs auto-lose is also unfair (but is arguably less unfair).

An example of poor seeding.

Break’s over…on to another topic. Let’s talk about seeding. I’ve been in a lot of Starcraft 2 tournaments, so seeding isn’t new to me. It makes sense. If the first round consists of about equally matched pairings, it’s unfair to the higher level players, who have a chance of losing while a lower level player has a chance of advancing. In the example above, the platinum level player gets by with a game against a silver player (near the lowest level), while the two masters players (the near highest level) are pit against each other.

There was an attempt at seeding by placing participants from the first tourney in later rounds, but absolutely no semblance of effort in the first round. As a result, first round matches were littered with both extremely close matches as well as complete one-sided roflstomping. Heck…use Alexa or something for all I care. As long as it’s standardized in some way.

I’ll end on something I noticed with the new voting system. The great new system that will make everything better has decided to put vote results up on the voting screen. Before, people would have to click to see results prior to voting, which means that it could be avoided unless you just wanted to see them first.

Now, they’re just there and will subconsciously affect the decisions of everyone, not just those that want to be influenced. The rich get richer, and it’s even more so with this new system. Arguably, this isn’t the biggest deal, but that would frankly be why I put it at the very end.

Well…that was quite a lot of stuff. For those of you able to make it through all of that information, congratulations. At the end of the day, most of this stuff is based on my perception, and I likely don’t know the full story (although am I really asking for much?). For example, I don’t know how they really approached seeding because it isn’t shown like the first tourney. Let me know if I missed anything…hope you didn’t find this too boring. And finally, good luck to anyone still in the tourney!

12 thoughts on “Aniblog Tourney Thoughts”

  1. Seeding: The seeds are based mainly on age and previous tourney score. We did outline that in the very first post. The reasoning was that older blogs have had plenty of time to advertise themselves so don’t need to participate in more rounds. As for why it seemed all over the place for the first round, very few blogs have many readers at all. Hence the blogs in the first round are near impossible to accurately predict who will beat who because they’re all so young (although I really should have just stuck Metanorn in the second round). Also I don’t get why people seem to want total accuracy with seeding. Is it really more interesting if all the seeds are exactly right? That ends with all the higher seeds automatically winning their matches against the lower seeds, which kills all tension. So long as it’s roughly correct, why do people care?

    Top vs bottom: I see that logic. Interestingly though, the blogs on top are always the ones that were originally seeded higher, so that higher percentage would be expected anyway.

    Too short polls: Last time the polls were open for a full week, but nearly all the votes came in the first 48 hours and remaining open for longer did next to nothing. Also the tourney ran for too long, so the long breaks as we waited for every match to end killed a lot of interest for people. Although I suppose the really long break we hadwhile we waited to get the new polling system killed interest anyway.

    Ignoring the rules: There were 160 blogs, and by my count there was a grand total of one that was a translation site that we didn’t pick up on, which got knocked out in its first match. Oh dear, the tourney is ruined -_-

    Voting system: Actually you were able to see the results if you clicked the ‘view results’ before you voted, although I get the psychological difference of not being instantly greeted with that.

    Hope that answered some of your questions


    1. Seeding: That’s sorta how a first round is supposed to be, though. A way to thin out the lower seeds. Sure, there are some times when the lower seed manages to make a comeback, which can be exciting, but “lacking tension” simply sounds like a crude justification.

      Too short polls: Could more matches be put into each week? I feel like that might help. Four blogs in two days seem easy to ignore, but something like ten blogs in a week seems to be fine.

      Ignoring the rules: Well, now that I know that it wasn’t intentional, it kinda goes with my point that more effort should be put into something with such a large scale. Just seems careless.

      Voting system: Like I tried to explain, in the earlier system, those who wanted to affect their results were given the opportunity to do so, while in the current system, everyone will automatically be influenced even if they don’t wish it to be so.


      1. We did use seeding though, including counting comments as D_S suggested below. We just didn’t publish the results, because we figured it didn’t matter. We didn’t publish the results last year either.

        I don’t know what you mean by ‘more matches in one week’. There were new polls up every day last time. 2 polls per day with 4 blogs on display. If we had used that same method this time it would have taken about 5 months to complete with the increased number of blogs competing. Last time people complained about the tourney being too long and was missing a bunch of blogs, so we included more blogs and shortened the time frame with each match. This time people complain about the total opposite. Perhaps you can understand my frustration here.

        As for ‘effort’, it’s one blog out of 160 we mistakingly thought passed. Please don’t say I needed more ‘effort’ put in because of one mistake


        1. Like I said, I understand that I don’t know the full story. I just want to point out that seeding should be consistent and as objective as possible.

          What I’m trying to say is that the more often you refresh, the easier polls get ignored. If every day, a new set of blogs appears, there’s little point in dwelling on other blogs, even if their poll lasts for a week. It’s basically a matter of spotlight. If you give more blogs the spotlight for a week, I would suspect they’d be less likely to be ignored than having fewer blogs in the spotlight for a day. I understand that you’re frustrated, and I’m sorry. Perhaps my suggestion was a bit too extreme, but I just see the two day time frame as too limited of a window. A visit to family or a bad cold and you completely miss the window.


        2. I actually think the staff did a fairly great job at seeding, with the clear exception of Metanorn and Agreed with staff that it’s a good effort.

          Further, the lop-sided results in round-one can’t be blamed solely on seeding. A lot of the lop-sided matches resulted from endorsements of certain blogs from another big traffic blog, but are not actually reflective of the participating blog’s own regular traffic. Another point of consideration. Two blogs may have similar traffic and comments, but the level of involvement of one blog’s reader may differ vastly from another. There is little to predict how committed readers are based on traffic alone. (Though comments are a decent indicator.) Another point, some authors simply have a better social media engine that isn’t directly related to the blog (Twitter, FB, G+, anime-stream forum thingy…). In addition, many lower-seed blogs share the same group of readers (fellow bloggers), so although on first glance, the blogs look similar in terms of traffic and comments, in reality, their shared readerbase may be completely skewed to one-side when one blog is more favored. Finally, campaigning differences. All these can make the matches in round-one lop-sided, close, and in general, seemingly inconsistent.

          I don’t know all the blogs, but for the little time I’ve been around getting acquainted with the blogosphere, the seeding looks about almost as good as one can make without precognition. That they’re not exactly 100% predictive is the reason we have a tournament.


  2. Your points totally make sense, but it’s not like anyone can get the organizing team to implement these suggestions. It’s pretty evident that (unlike last time) the organizers themselves aren’t too enthusiastic about the whole thing. Frankly, it seems they’re running the tourney this year not because they care, but because they felt obligated by prior promises. It’s nice that they at least finally got the new voting system together, but overall they seem more interested in making excuses and saving face than attempting solutions. Not that I’m saying anything new to you here, but I wanted to throw in a bit of support.


    1. Even if they don’t get implemented or they get completely ignored, these felt like things that I just had to say. With so many posts complaining about the popularity contest or whatever, I feel like the more technical points that people ignore should be more important because of the fact that they can actually be fixed.


  3. I don’t know if I’d use Alexa to determine seeds. A better measure would be a cursory glance at the number of comments that articles receive. Even if Alexa’s ratings were accurate, a high readership isn’t necessarily a loyal readership, and loyalty is generally what wins these polls.


    1. All I’m looking for is something consistent and as objective as possible…doesn’t matter to me what it is. Alexa was just the first thing that came to mind.


  4. I’m curious as to your thoughts on the overall final winner. True victory or a sham of a champion? With ballot stuffing and similar complaints, should the last blog standing be considered illegitimate? Maybe put a little * by his record…


    1. First off, I don’t really see ballot stuffing as a problem. Votes are votes…however you get them. At the end of the day, one person has to win and it’s not easy to be that one person, so I respect that person’s ability. It’s never luck that gets them there. I wouldn’t be so foolish as to try to devalue his/her victory.


Leave your comments here

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: