I'm mostly done compiling the lists for this year, and about to begin making preparations for the results threads (coming up soon!), and...

I kinda wanna make a question to y'all. I've been thinking of changing up the scoring system for a while now, but never really took it seriously because I wanted to keep it consistent throughout the decade. We're past the first decade now, though, and I've been giving this a lot of thought. I'm experimenting with median-based rankings and I've been surprised by how much it actually shakes up the results, but it's hard to find the right balance, where to draw the line before things start getting too arbitrary. I don't want it to ever feel like I'm personally influencing the results, but I'm interested in the idea of changing up the system as it's led to very repetitive lists over the years... I mean, at this point anyone could just make a fake mockup ranking of VGC's favorite games and it'd be just as easy to believe as the real thing. These lists are so predictable

Anyways, trying to keeping the question simple here.

Are you interested in a reworked scoring system, or do you feel things should be kept as they are?

Edit: also, the Official Thread will be locked up in a bit as I'm done compiling and will not receive any further lists from now on.

Be careful in reworkign the scoring, but some change would not be wrong. As I see it, the goal of a scoring system for this should be:

• boost games that often make a list
• boost games that reach high positions

The current system does that kinda already. At least I think it does. The current system is something along the line of 1 point for #100 and 100 points for #1 and adding it all up, right?

Median is very risky. Someone may have an obscure game he likes very much near the top of his list, but nobody else has the game (because nobody else played it). Another game might be well known, well liked and often played by many, so it enters a lot of lists, but not always at the top. The second game has a lower median than the first, and that is something you probably don't want.

I understand that many are frustrated seeing the same games everyone knows and kinda likes at the top, while more obscure games have no chance. But instead of using a median, I would simply give some bonus points for top-placements. Say 1 point extra for TOP 50 games, 2 points for TOP 25, 3 points for TOP 10, 5 points for TOP 5. This would help elevate the games reaching top-placement in some lists rise above often named but lower ranked games. But be careful not to overdo it.

I'm mostly done compiling the lists for this year, and about to begin making preparations for the results threads (coming up soon!), and...

I kinda wanna make a question to y'all. I've been thinking of changing up the scoring system for a while now, but never really took it seriously because I wanted to keep it consistent throughout the decade. We're past the first decade now, though, and I've been giving this a lot of thought. I'm experimenting with median-based rankings and I've been surprised by how much it actually shakes up the results, but it's hard to find the right balance, where to draw the line before things start getting too arbitrary. I don't want it to ever feel like I'm personally influencing the results, but I'm interested in the idea of changing up the system as it's led to very repetitive lists over the years... I mean, at this point anyone could just make a fake mockup ranking of VGC's favorite games and it'd be just as easy to believe as the real thing. These lists are so predictable

Anyways, trying to keeping the question simple here.

Are you interested in a reworked scoring system, or do you feel things should be kept as they are?

Edit: also, the Official Thread will be locked up in a bit as I'm done compiling and will not receive any further lists from now on.

I do think there are problems with the rankings and I have a couple of suggestions at fixes/changes.  But I first want to say that predictability is not a bad thing in and of itself.

1) Third party games do not score high enough. I think the fix here is simple in principle, but I don't know how much work it would put on you.  The simple fix is to automatically combine every possible platform into their score.  Right now GTA5 could be scored on 5 different possible platforms and next year it will be 7.  When the score is split among 5 platforms then this type of game really gets short changed.

2) Mario and Zelda dominate the top slots too much.  Most people end up really loving the Mario and/or Zelda game that was popular when they were a kid.  I really like that many different generations of gamers are participating, but this results in lots of Mario and Zelda games in the top 20, because almost every major entry is there.  I say this as someone who put a Zelda game as their #1 and a Mario game at #6 from the same platform (the popular Mario and Zelda from when I was a kid).  Clearly all of these games deserve to be recognized to some extent, and that is why they are so loved, but maybe not so many Marios and Zeldas deserve to be in the top 20.

My suggested fix is to give all games more points just for making it into anyone's top 50.  IIRC last years method was that a #1 game got 60 points and a #50 game got 11 pts.  Is that correct?  I would try something more like #1 game gets 100 pts and #50 game gets 51 pts.  Or if you want to be even more extreme then #1 game gets 150 pts and #50 game gets 101 pts.  I think this will dilute some of the Mario and Zelda problem.  I don't think all of these games are actually appearing more frequently on lists than every other game out there.  Instead, I think most people have their favorite Mario and Zelda and rank them really highly.

If you make these two changes then I think it will fix the problems with the lists.  Games like GTA5, Minecraft, Dark Souls, etc... will end up moving higher in the rankings.  I suspect there are several third party games that are actually popular with the community here, but they have been ending up with a worse rank just because of how the scoring system is structured.

EDIT: I just want to add, that in light of everything I've just said, there is a lot about the current system to be praised.  One of my pet peeves with other "Top game of all time" lists is when most of the top games were released in the past 10-15 years.  Another thing that irritates me is when 1-2 consoles are represented far more than all of the others.  I don't think the VGChartz system falls into either of these traps.  In making changes we want to hold on to the parts that are good while fixing the parts that are bad.

This definitely seems more reasonable than what I thought and when you think about it it's not so different from the old system, assuming I've understood it right. Before it was the sum which is also equal to average*number of votes. Now instead it's median*number of votes.

The difference would be that games than get most high rankings and a few lower ones would get more points since their median is higher than their average and the opposite for games that get mostly lower rankings and a few high ones. Games that get ranked #1 on a single list should get the same number of points as before though, so I'm not sure I necessarily see that as a problem. I would assume they still wouldn't place too high.

This definitely seems more reasonable than what I thought and when you think about it it's not so different from the old system, assuming I've understood it right. Before it was the sum which is also equal to average*number of votes. Now instead it's median*number of votes.

The difference would be that games than get most high rankings and a few lower ones would get more points since their median is higher than their average and the opposite for games that get mostly lower rankings and a few high ones. Games that get ranked #1 on a single list should get the same number of points as before though, so I'm not sure I necessarily see that as a problem. I would assume they still wouldn't place too high.