I don't understand site rankings?

I don't understand how this site ranks the "Editor's Pick" and "Most Popular" categories. Worm, for example, is ranked below Tales of MU in both lists, despite the former having higher ratings, more bookmarks and twice the number of recommendations. Is this a website error, perhaps? If not, could someone explain the formula used to calculate and sort these web serials?

Worm was at the top, but completed serials rank lower than currently active serials.

Ahh, I see. I got four comments tonight about that, sent an email to the 'feedback' email linked in the 'info and submissions' page. I'd say that makes sense, but it doesn't really make sense to me. Oh well. Thanks for clarifying, Jim.

I occasionally wonder if there should be a "completed serials" list that's as easily accessible as the "Editors Picks" or "Popular" lists. I suspect there are people who would exclusively go for completed serials given the opportunity.

I suspect there are people who would exclusively go for completed serials given the opportunity.

Oh, there is totally a demographic for this. It would be a great idea. :)

There's a complex formula of weightings, including such considerations as age of the listing and freshness of the votes, as well as current update status and frequency of updates. It's supposed to allow for turnover, make it easier for new serials to make their way to the front pages.

http://webfictionguide.com/complete-novels/ and http://webfictionguide.com/complete/

I can see there are issues with those listings, too. The aging formula needs to be changed to age listings from the time they are finished, not the time they are listed. Unfortunately, I don't have time to fix it right now. It's on my to-do list.


Fiona, Jim: if either of you want to create a "complete webserials" tag and apply it to some of the listings, please go ahead and do it. I'll put it on my list, but can't make any promises about when I'll be able to get to it.

Okay, I hacked it a bit to fix it for now: I've set the listing dates for Worm and Street to the date they finished. Ugly, but will do for now. If you notice any other serials that should be updated so, let me know.


I'm personally of the opinion that a completed story should be ranked higher than an ongoing one. A complete story can be rated for its entire structure - a work in progress is not yet the author's full vision. An ongoing story hasn't fully realized its theme or plot - and an ongoing one can go off the rails.

It took me years to give Legion of Nothing 5 stars because it is an ongoing story - but it has maintained a consistent level of quality for so long that eventually I had to note the achievement. It is the Lou Gehrig or Cal Ripken Jr of serials I have read. On the other hand, Tales of MU has been around just as long and over time I have taken stars away because the story seemed to meander and repetition led to me perceiving a lack of quality over time.

A complete story is static and certain - you can review the entire work for its full impact. I think writers who have told a complete story deserve to have that recognized. Newer is not always better.

I see Gavin's point. On the other hand, the primary form of fiction covered on WFG seems to be serials. Automatically downgrading a story actively participating in the primary form of fiction covered on the site seems to work at cross purposes. It's like comparing a TV show to a movie: a movie *can* be evaluated as a whole. A TV show will be an amalgam of all its episodes to date, perhaps skewed in favor of the good, since a lot of people invested in actually following a show will tend to err on the side of liking it.

Worm is still a serial even if it is finished. I don't see why it should be downgraded for achieving completion, when that is an accomplishment. Same goes for any other story, just Worm is a recent example.

I'm with Gavin. I'm sure there are a lot of people who enjoy the ongoing serial aspect, but it makes no sense to me that completed stories get shuffled into a corner by comparison. They should be given equal screen-time as ongoing stories -- the idea of putting them in different sections seems like a good one. Then readers can make their own decisions rather than being heavily weighted in one direction.

I don't think it should be downgraded--if I implied that, it wasn't intentional. Your argument was that completed works should automatically be ranked higher because they're complete. That's the point I disagree with.

I feel sort of awkward about voicing my own thoughts, because of how the conversation got started. That aside, I do want to stress that I have the utmost respect for Chris and the work he's done on the site. I owe WFG a lot, as it got me much of my initial audience as well as contacts I maintain today.

I guess what I'm going to say is that there's an unfortunate stigma against serials as a whole because serials often don't finish. As I saw recommendations for Worm pop up on Reddit and in forums, I saw responses saying that people weren't interested because they didn't want to get invested in something that would taper off or go on indefinite hiatus, which serials supposedly do. My desire to combat and change this notion was a kind of encouragement as I reached the final months and weeks.

For every serial that's abandoned or serials that stop, I think we all suffer, just a tiny bit. It's less readers, and just as the proverb goes, 'it takes money to make money', it takes readers to get readers. For every good work out there, we create more readers, and for every work that finishes, we change a few minds and we create more readers for serials as a whole. I am immensely grateful to WFG for championing good works and helping people to find them (and I was even before I started Worm, and before Worm was considered good by anyone), and I think the next logical step is to change the context where we're talking about completed serials.

Rather than have 'Popular Serials' and then 'Completed Serials' as if they're an exception or a side note (requiring multiple clicks to find), I think it makes sense to make the two binary/equally visible. With the metrics being somewhat opaque, in terms of how things are ranked (ie. what is 'popular'? What is 'full list'? Why are they different?) perhaps make it more clear cut?

An example would be:

Completed Serials: Highest Rated

Completed Serials: Trending

Ongoing Serials: Highest Rated

Ongoing Serials: Trending.

I know that, as you said, Chris, it's hard to find time to work in this or any related department, so it's just food for thought/a topic for discussion. Given how much I've gained from WFG, I'd be willing to donate - I know having the spare change to buy fast food/something easy to prepare has made the difference in my being able to find time to write vs. having to cook - maybe it could be the same for coding? Or perhaps it'd help cover site costs?

Oops sorry I should be clearer with my terms. Right now it sounds like star ratings are sorted with weighting favouring continuing stories over completed ones. I personally would weight the other way because you're viewing something complete versus a work in progress that hasn't reached its perfect or imperfect conclusion. Ranking is different in that a 2 star story that's complete is just as bad as a 2 star story that's ongoing.

I would be cool with equal weighting because that sounds fair - but it is easier to say Friends was a great tv series because it is done. How I met your Mother might go off the deep end in its last season.

That said, there's nothing about the WFG software I'm particularly happy with, 5(?) years in. Unfortunately, work is just so busy right now that I don't have the time to really tackle rewriting it. Maybe in January I'll be able to do something about it.

The current voting system for TWF is also easily gamed without the time decay. Anyone with a high "enough" vote count can keep topping off their votes, basically keeping other stories from appearing if a fanbase wants to keep at it.

To "Sock" -- the Editor's pick is not a rank. It is a collection of RSS feeds or manually updated feeds. To be included, an editor has to rate the story a certain level or higher. No "ranking" there. It's just based on update schedule of some sort.If a lot of those picks are updating, the RSS feed goes quickly. YOur favorite is going to get pushed down. That's the way things roll.

How does one become an editor's pick? Currently Palladian and Fiona are more active but there are a whole host of others who might select a serial to be one. Some of the past editors haven't been active in a while (although a few stop by once in a while.)

Regarding the rankings -- WFG offers multiple means of helping people find stories. There is the review system highlighting current review on the front page. There is the aforementioned Editor's pick. Then there is the lesser known novelsonline.info that lists by overall ratings irrespective of update schedule or complete/non-complete status. (This is a good keyword match to "online novel," which I have religiously been querying since starting to blog about serials.) ANd then there's Top Webfiction which is time-based but offers multiple time periods for the TWF listings.

There is also the complete tag as well.

WFG is not the only directory to use somewhat mysterious/inscrutable systems to push exposure around more equitably. Muses Success uses some kind of weighted mean/time formula to deal with story ranks. (I think most stories start with a mean score of 5 and have to rise beyond it.)

Wattpad's ranking system is also time based, likely for similar reasons to Chris'. More frequent changes to it positively affects current writers' motivations and also provides fresh exposure to newer stories in the mix. The past high-performers are found through other aspects of their system -- namely sorting by "most read" stories.

Of course, all their metrics are internally housed so they can make the determination of "most read." I think the equivalent on this site isn't quite available but perhaps someone can come up with another easy to implement "view" that Chris and editors would consider versus advocating an overhaul to existing algorithms.

I concurr with Chris' analysis for coding the site the way it is. To allow serials to fall by the way side as they are completed so that the new blood can be brought forward and fawned over the same as others. This may sound a bit selfish on my part since I'm in the "up n comming" category; but I assure everyone that is not the case and I see what the vision was for this website.

I, like wildbow, applaud Chris' efforts with this site and the benefits that come from getting listed here. I am finding a lot of visitors actually reading every single chapter when they come by my site. This is encouraging because it means I am capturing some of the audience and bringing them back.


Instead of manipulating the dates and stuff for completed serials, would it be easier to set up a "Greatest Hits" page for those completed serials that have garnerd the respect and admiration of their readers? Maybe base it on most favorable reviews and votes? I apologize if this is a stupid idea as I'm not antiquated with the programming aspect of this site.

I think we can all agree that it's not fair for any one group of stories to hog the limelight. I certainly don't want to stifle new serials. More that I'd like equal representation under the code, if possible and when feasible. The current weighting has an effect on the landscape -- going by my traffic figures, the referrals I get from WFG have dropped by about 80% since STREET was listed as complete.