Skip to main content

On Sale: GamesAssetsToolsTabletopComics
Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
TagsGame Engines
(2 edits) (+3)

To be honest, I find it hard to believe that there haven't been any changes to the algorithm or some kind of mass flagging or something. From November 5th to 8th we've lost 70% of our traffic, and I've seen complaints from many developers about the same period.

And ever since we continue to drop in the Popular lists for tags. Around Nov 5th, we were on the fourth row in Popular for our main tag, and today we're in 3000th place in Popular for the same tag, even though the traffic to the page has stabilized.

I understand that the algorithm needs to be opaque to prevent exploitation, but it would be nice to at least know the big no-no's that could cause this kind of fall.

I do not think it is any kind of flagging, since, the tags still exist and are populated by the same games. So if there was a "flaggable" problem, why just garble the ranking. There is no benefit. Also, if we assume there are the same number of players browsing, there should be an unexpected increase in views for other games.

The change many speculate about could be a change in weighting of recurring visitors vs new visitors. There is a difference between a user already following you visiting your game page and a new visitor looking at the game. It is a small data sample, but without considering the game in the OP, three of the games complained about here are web games. Those might have lots of recurring visitors.

Anyways, even small changes could have a runaway feedback effect for some games. Especially if some of the fans were located in the regions with the connection problems I read about in other threads. 

(+1)

I think the pill would be somehow easier to swallow if everybody was affected equally, but I can't figure a valid explanation for why some games at the top of the "popular" listings were unaffected, despite featuring the same themes, tags, etc as other games that were affected. This does suggest some kind of mass flagging. And if there was, my question is whether this was a one time rebalance that can be reverted based on how popular the game is in the following months, or a permanent flagging, which would hinder the game's visibility no matter how much effort we put into it and how popular it is with users.

(1 edit) (+1)

Initially I thought it had to do with recent updates. Even the game of OP had an update when it recovered. The games on top of the list of some tags had updates no older than a week or so, when I checked. But there also were too many old games.

But whatever it was, games seem to recover to the adjustment or the bug.

around place 110 of 725

That game now is around place 75 of 726.

I have trouble believing in any kind of flagging, if games can simply recover from it. Also, I do not believe there is any kind of intentional flagging. If they "flag" anything, they de-index it. But I fully believe in the possibility that there could be circumstances in wich the algorithm would decrease the ranking. This includes missing number due to connection problems, adjustment of weighting for recurring users of a web game, random alotment of the boosts it gives away to rotate games, adjustment of weighting of minor/major updates, false triggering of some anti-ranking-cheat, correction of previuosly false attributed view count and of course, bugs.