The rating system is fine. In fact, even though they allow half stars, I only use full stars in my user reviews. It's a good system.
PLEASE update the rating scale!!! 80 or 100 is not appropriate.
@pw2566ch said:
@RollingZeppelin said:
@pw2566ch said:
@Capt_Ventris said:
@pw2566ch said:
@Pinworm45 said:
The five star system is the best system for rating video games.
Do not think of it as 80% because that is silly.
I disagree. If I had to choose any rating system. It would be a 10-scale system. Or what the OP was mentioning about .5's and such. It seems like the 5-scale rating system is best because most reviewers don't follow it right. As far as I know, Jim Sterling is the only one.
I do not understand what the difference between a 7-8 would mean in quality. Let alone the difference between 7.0 and 7.5.
The five star system works so well in clearly giving an indication of the quality of the game in the reviewers mind.
The 1-10 scale is more detailed. I understand the 5 point scale is a much simpler aspect, but the 10 point scale offers reviewers more room to work with. For example What if the game has game changing bugs in the multiplayer, but you know that the single player is perfect. You know that giving it a 3/5 is discrediting the game, but giving it a 4/5 feels like you're not warning the consumers enough. That's why the 10 point scale would work perfect in this case.
The only thing that's wrong with the 10-point scale is that most reviewers abuse it. Mainly most publishers tend to hound review sites if they notice that their game received between a 5-7. There's even cases that publishers will pull early copies from the review site and review sites don't want this. So, usually they'll give games higher scores than what is necessary. I believe this article sums it up just about right.
Like the OP you are completely missing the point of GB's scoring system. It's not 3/5 or 4/5 is 3 stars or 4 stars, the difference is qualitative vs quantitative. They are not percentages, and the issues of your example would be talked about in the actual text of the review. If someone bases their purchasing decisions for entertainment based solely on a number then they are an idiot.
Hell, the title of the article you referenced is "Further Proof Quantified Rating Systems Are Flawed", and you use this to argue that a 10 point quantitative rating system is better than a 5 point one.
*facepalm
I don't think you understand what I'm saying. I know people should read the review to make their purchasing decision, but most people don't. They look at the score and buy it based on that. Why do you think publishers are so hard up on reviewers? I also never said that the 10 point review system is flawed, but it's not perfect either. If publishers would get off of the reviewers ass about reviews or if reviewers would stop worrying about early copies of games, then the 10 point score would be perfect. It's a simple fix.
So could you chill out for a second? This is only my opinion. I'm not telling anyone that this is how the gods wanted it, so let it be done.
You still don't understand. The 10 point score is a quantitative system, it is flawed, and it fixes absolutely nothing, which is clearly argued about in that article. I'm not trying to get at you, but do you understand the difference between a quantitative and qualitative metric?
If there is no real numbered score then the people who do read a score and buy a game based on it will be forced to read the review.
I'm just trying to point out the flaws in your logic, I'm not attacking you. Think critically about what I wrote and address my argument directly.
@TaliciaDragonsong said:
As I said before. Fuck Metacritic.
Yes, I always use gamerankins for my avarage score info. It's important to be able to distinguish a 89.97 game and a 89.98 game ^^
Seriously, the 5 star scale is the best, it tells you what you need to know. We have to get away from seeing 4 stars as 80/100. 4 stars is 4 stars out of 5, nothing else.
I talked to a reviewer about a decade ago that used the 0-100 score system and asked him what the difference between a 89 and a 90 was and why is latest review was given 89. His answer was that he had already given out several 90 scores recently so he didn't want to give this game the same score. I thought that was really stupid, but also very honest since who can tell the difference of 1% anyway? I think a 10 grade scale without any half points and such is about as far as you could stretch it and still make sense, but 5 stars says enough. No one reviews films with a 100 grade scale or even a 10 grade scale. Film people always used to laugh at me when I said a game was given 74 since he was used to 1-5. Not to mention that in the 0-100 scale 75 seemed to be avarage. I remember in the late 90's a magazine thwt I bought and the worst game had a score of 68/100 and according to the review that was pretty much crap. It was like the 0-50 on the scale was never even used so why even have it? Which the realized around 1999/2000.
i say lets use a 2 point scale:play it or dont. read the review, and watch videos. then try it yourself. what do you care if mario didnt get the score you agree with, people are allowed to have differing opinions.
its something theyve talked about before and have thought it through. Theyve been through all different kinds of point scales. They want the star rating, they find it to be a much more comfortable scale than conforming to the 10 and 100 point scales. Especially when you consider the gamespot review system where they had to use decimals as well for a long time before they left
I actually disagree the rating system on giant bomb isn't much better than on any other site. I'm pleased they don't give each game a finely detailed score but scores in general are stupid regardless of if they are out of 5 or 100. I would say tags or short descriptors would be better suited. I know this was attempted on gamespot and never really worked but i think it tells people more about the game at a quick glance but really people should just read the review to find out more about the quality of the game.
The idea of using descriptors would be tagging a game as difficult for instance. Some people would see that as a main selling point and others would see it as a reason to avoid the game. This is the kind of thing a star rating can never tell you.
The number of stars do not translate into percentages, I believe the translation is as follows;Thank you!
5/5 - PLAY THIS GAME!!!!
4/5 - This game is very good, we would recommend you play it.
3/5 - This game is good, but has some problems.
2/5 - Don't play this game unless you're a fan of the genre/series or can overlook the numerous problems.
1/5 - Do not play this game.
Why can't people understand this scale? It's perfect. Joystiq uses this scale too.
The 100 point scale is dumb unless reviewers make 50=average. Games should go up or down from the middle. The current 100 point scale 70 seems to be average so you have a system where better games can gain 30 extra points. Terrible games can lose 70 points yet it rarely gets used. The Beer Mag I get uses the damn 100 point scale then every damn beer is between 70% and 100. 5 stars as described above is the best system. The only other number system I liked was IGN old way of rating each aspect of a game( sound, gameplay, controls, graphics, etc etc.
No.
It is fine the way it is.
Even then, I don't care for review scores especially anything involving .1-.9
This is what the ratings systems tell me:
5/5(90-100) - We recommend you play it but there is a chance you may not like it as much as we did.
4/5(75-89) - It's very good but has some problems.
3/5(60-70) - It has more bad things than good things.
2/5(below 60) - Game is bad. Don't play it.
1/5 - Kill it with fire.
@delt31 said:
I love this website. I support it with a membership but the rating scale is unnecessarily harsh. Four out of Five or Five out of Five (80 vs 100) is not appropriate. There needs to be a 90 or something in between. I'm well aware that the guys on the site want the reviews to answer whether you should buy it or not but cmon - mario super land 3d as an 80? I believe even Jeff would give it a 4.5 out of 5 or something like that.
Please consider redoing the scale.
Thanks!
It isn't GiantBomb's fault metacritic translates their scores that way. The 1-5 star ratings are for easy understanding by readers not for some developers bonus. Besides, what is the difference between a game that scores an 82% vs an 84%?
@RollingZeppelin said:
@pw2566ch said:
@RollingZeppelin said:
@pw2566ch said:
@Capt_Ventris said:
@pw2566ch said:
@Pinworm45 said:
The five star system is the best system for rating video games.
Do not think of it as 80% because that is silly.
I disagree. If I had to choose any rating system. It would be a 10-scale system. Or what the OP was mentioning about .5's and such. It seems like the 5-scale rating system is best because most reviewers don't follow it right. As far as I know, Jim Sterling is the only one.
I do not understand what the difference between a 7-8 would mean in quality. Let alone the difference between 7.0 and 7.5.
The five star system works so well in clearly giving an indication of the quality of the game in the reviewers mind.
The 1-10 scale is more detailed. I understand the 5 point scale is a much simpler aspect, but the 10 point scale offers reviewers more room to work with. For example What if the game has game changing bugs in the multiplayer, but you know that the single player is perfect. You know that giving it a 3/5 is discrediting the game, but giving it a 4/5 feels like you're not warning the consumers enough. That's why the 10 point scale would work perfect in this case.
The only thing that's wrong with the 10-point scale is that most reviewers abuse it. Mainly most publishers tend to hound review sites if they notice that their game received between a 5-7. There's even cases that publishers will pull early copies from the review site and review sites don't want this. So, usually they'll give games higher scores than what is necessary. I believe this article sums it up just about right.
Like the OP you are completely missing the point of GB's scoring system. It's not 3/5 or 4/5 is 3 stars or 4 stars, the difference is qualitative vs quantitative. They are not percentages, and the issues of your example would be talked about in the actual text of the review. If someone bases their purchasing decisions for entertainment based solely on a number then they are an idiot.
Hell, the title of the article you referenced is "Further Proof Quantified Rating Systems Are Flawed", and you use this to argue that a 10 point quantitative rating system is better than a 5 point one.
*facepalm
I don't think you understand what I'm saying. I know people should read the review to make their purchasing decision, but most people don't. They look at the score and buy it based on that. Why do you think publishers are so hard up on reviewers? I also never said that the 10 point review system is flawed, but it's not perfect either. If publishers would get off of the reviewers ass about reviews or if reviewers would stop worrying about early copies of games, then the 10 point score would be perfect. It's a simple fix.
So could you chill out for a second? This is only my opinion. I'm not telling anyone that this is how the gods wanted it, so let it be done.
You still don't understand. The 10 point score is a quantitative system, it is flawed, and it fixes absolutely nothing, which is clearly argued about in that article. I'm not trying to get at you, but do you understand the difference between a quantitative and qualitative metric?
If there is no real numbered score then the people who do read a score and buy a game based on it will be forced to read the review.
I'm just trying to point out the flaws in your logic, I'm not attacking you. Think critically about what I wrote and address my argument directly.
Gentleman I propose a race around the world
Whatever I need a "0.00-100.00" rating system. I don't buy games that are under 74.56 rating, anything under that is such a waste of time!! Also who reads anyways, I didn't even read this thread but the title. Reading the review doesn't give me a number valve, the wall of text just ends up confusing me. I got it, What I should do is have someone read the review for me then break it down in to the 0.00-100.00 rating system I want!!! YES I'M A GENIUS!!! PROBLEM SOLVED!! you can lock and or delete this thread now.
I think they need to set up a new scoring system based on all games they have ever played. The top score will thus be different for each reviewer based on the total number of games he/she has played in their entire lifetime, the top score will go up 1 with each new game played, and all games will be re-scored as needed to keep them in their proper place with regard to all other games.
Example: Ryan has played 4321 games and starts on Assassins Creed: Revelations. His top score now becomes 4322. He determines that AC:R is 3471/4322. When this score is assigned, the current holder of that spot (let's say it was Dead Rising: Case 0) and all games above it move up 1 point. This way there is never any argument as to which game is better than which other game because you have a master list from each reviewer to look at.
When was this turning point in history, where metacritic became a reference for anything?
I mean, really, one day I woke up, and heard about the first time, went out to the depths
of the net to have look for it, and thought to myself, oh, what a place of obscurity,
never visiting it again.
The larger the range of a scale, the less it may really say about it's subject.
Articles, comments, and quicklooks, they may tell you something, in
combination with an easy to put in context score system, like 1to5 stars.
There were some well written, critical articles, about games with a
mediocre score, where I thought myself, thank you Mr. Reviewer,
the things you criticize, are exactly the old-school stuff I like.
Sometimes you see those two or three stars, and think,
huh, what's that all about, I don't have this on a numeric
scale from 1 to a thousand gazillion.
@pw2566ch said:
@Capt_Ventris said:
@pw2566ch said:
@Pinworm45 said:
The five star system is the best system for rating video games.
Do not think of it as 80% because that is silly.
I disagree. If I had to choose any rating system. It would be a 10-scale system. Or what the OP was mentioning about .5's and such. It seems like the 5-scale rating system is best because most reviewers don't follow it right. As far as I know, Jim Sterling is the only one.
I do not understand what the difference between a 7-8 would mean in quality. Let alone the difference between 7.0 and 7.5.
The five star system works so well in clearly giving an indication of the quality of the game in the reviewers mind.
The 1-10 scale is more detailed. I understand the 5 point scale is a much simpler aspect, but the 10 point scale offers reviewers more room to work with. For example What if the game has game changing bugs in the multiplayer, but you know that the single player is perfect. You know that giving it a 3/5 is discrediting the game, but giving it a 4/5 feels like you're not warning the consumers enough. That's why the 10 point scale would work perfect in this case.
The only thing that's wrong with the 10-point scale is that most reviewers abuse it. Mainly most publishers tend to hound review sites if they notice that their game received between a 5-7. There's even cases that publishers will pull early copies from the review site and review sites don't want this. So, usually they'll give games higher scores than what is necessary. I believe this article sums it up just about right.
I'm curious, how exactly WOULD the 10 point scale work perfect? 6.0 vs 7.0 doesn't really tell me anything, neither does 7.0 vs 8.0, there's no real defining difference, either one could be taken as 'discrediting the game' or 'not warning consumers enough'.
@Inkerman said:
The number of stars do not translate into percentages, I believe the translation is as follows; 5/5 - PLAY THIS GAME!!!! 4/5 - This game is very good, we would recommend you play it. 3/5 - This game is good, but has some problems. 2/5 - Don't play this game unless you're a fan of the genre/series or can overlook the numerous problems. 1/5 - Do not play this game.
thank you. so many people don't that Giant Bombs review scale is not numerical, but instead a recommendation scale.
@delt31:
I gotta agree, but mostly just because sites like MetaCritic insist on altering a 4 out of 5 to an 80 out of 100. The easiest and most obvious adjustment is to move to 10 point scale with not .5s, since .5 seems to be the deal breaker rather than 10 points.
1 to 5 stars is stupid. We should review in terms of other astronomical bodies, such as planets. Asteroids would obviously get an asteroid. Sonic would get a Saturn because it has rings and shit.
@ImHungry said:
@pw2566ch said:
@Capt_Ventris said:
@pw2566ch said:
@Pinworm45 said:
The five star system is the best system for rating video games.
Do not think of it as 80% because that is silly.
I disagree. If I had to choose any rating system. It would be a 10-scale system. Or what the OP was mentioning about .5's and such. It seems like the 5-scale rating system is best because most reviewers don't follow it right. As far as I know, Jim Sterling is the only one.
I do not understand what the difference between a 7-8 would mean in quality. Let alone the difference between 7.0 and 7.5.
The five star system works so well in clearly giving an indication of the quality of the game in the reviewers mind.
The 1-10 scale is more detailed. I understand the 5 point scale is a much simpler aspect, but the 10 point scale offers reviewers more room to work with. For example What if the game has game changing bugs in the multiplayer, but you know that the single player is perfect. You know that giving it a 3/5 is discrediting the game, but giving it a 4/5 feels like you're not warning the consumers enough. That's why the 10 point scale would work perfect in this case.
The only thing that's wrong with the 10-point scale is that most reviewers abuse it. Mainly most publishers tend to hound review sites if they notice that their game received between a 5-7. There's even cases that publishers will pull early copies from the review site and review sites don't want this. So, usually they'll give games higher scores than what is necessary. I believe this article sums it up just about right.
I'm curious, how exactly WOULD the 10 point scale work perfect? 6.0 vs 7.0 doesn't really tell me anything, neither does 7.0 vs 8.0, there's no real defining difference, either one could be taken as 'discrediting the game' or 'not warning consumers enough'.
One to five is simple and to the point. Why get more granular and open this basic system to the fallacy of a 10 point system. We all know that anything under a 6 (maybe 7) isn't worth your money, so already over 50% of the scale is lost. At least here a two star game can have redeeming qualities... not many but some. And a five star game doesn't have to be "perfect" but can still get top marks...
@Pinworm45: Jim Sterling is being held up a good example of a critic/reporter? Jim Sterling? We have some very different opinions.
@pw2566ch said:
Actually, I think everyone should throw away all review scores. Just post the review. No score. It will literally get rid of Metacritic (the gaming portion) and throw off the publishers completely.
If only.Unfortunately a lot of traffic to game sites comes from Metacritic. Noone would be brave enough to cast them off by themselves just yet unless everyone did so.
If anything, abolish review scores. But that won't happen, so eh.
Even then, I don't remember the last review I read. I find them to be pretty useless, I know what games I want and don't want myself.
@delt31 said:
I'm well aware that the guys on the site want the reviews to answer whether you should buy it or not
You answered your own question. What's the difference to you if something gets an 80 or a 90? What does that really mean? You know you're getting 3D Land already, and the 4/5 means that if you're at all interested, yes, you should get the game. A 90, means what? An 80 means that if you're interested in this type of game, get it. A 100 means that if you're interested in games, and have even the slightest passing interest in the genre, get it. So what would a 90 accomplish? Nothing. It doesn't matter if it's harsh or not. It doesn't matter if it "messes up" metacritic (which is inherently messed up), it just matters if you should buy the game.
The scale should be based on how many bullets you'd put into your revolver before starting your after-game session of Russian Roulette. BF3? Great multiplayer, poor single player. One bullet. Rayman Origins? A delight, no bullets. NFS: The Run? Serious flaws, five bullets. Sonic 2006? Six bullets.
And for those of you who think six bullets is too close to the limited freedom of five stars, you can change the caliber of gun. .44 Magnum for Big Rigs, etc.
Normalize that, Metacritic!
You act as if it matters. Review scores might as well be a yes, no, or maybe. Any more detail than that and people start obsessing about pointless numbers like the difference between and 8 an a 9. Much like you are right now. The 5 star system is the best method that I've seen used, and you shouldn't be trying to convert it up to the 10 point system to get your score.
I think a 5 star or a 1up style A - F scale is better. The difference between say a 5 and a 7 when actually reading the text of reviews -- regardless of site -- isn't really noticeable. I understand that a 100 point scale offers more granularity, but it doesn't seem to be utilized by most writers. Only games that are noticeably buggy or broken or just an unfinished mess seem to be rated at less than a 6.0 (60) on most 100 scale sites. There are exceptions (Jim Sterling) who use the entire range of scores, but most people just think of those writers as trolls.
Well I can understand why someone would want to change the system. I like both the 5 and the 10 scale system with and without decimals. I like the 5 star scale because I need to think for a long time about what score I want to give a game. A 10 point scale, especially with decimals, makes each step up or down less meaningful. That being said, I often find that a score from a 10 point scale gives a more "accurate" picture of the content of my review. I also understand why Jeff likes the 5 star system, and I think they should stick with it. Changing the system for metacritic is a stupid idea.
One of my personal favorite things about the five star scale is that it actually forces me to really think about how the score and the text line up. I've stopped giving scores until I read over the final submission draft and am satisfied with it. Without those nebulous "in-between" numbers, it forces me to really think about whether or not the score in my head makes sense for the experience I'm describing. If anything, I think it's actually made my reviews better.
So, no, I don't think we'll be altering our review scale any time soon. If this is being suggested because it somehow applies to Metascores or whatever other similar horseshit, no. Just no. We absolutely, positively do not give a rank shit about how our scores are portrayed on Metacritic.
They don't use a 100 point scale. They use a 5 star scale. Mario did not get 80 it got 4. Does it really change anything if it had gotten an 83.247/100. Tying a number to an abstract opinion is pretty silly anyway. If they were to do anything to their rating system I'd hope they would go the other way and use a Siskel and Ebert inspired review system.
The OP obviously has not studied statistics, otherwise he wouldn't be calling the rating system illogical since while most review score systems are based on ordinal scales (i.e. 10 is better than 9 is better than 8 is better than 7, etc), the GB 5-star system seems closer to nominal scaling since the review scores are really categories, as the purpose is not to say that a 5-star game is "better" than a 4-star game but rather a 5-star game can be recommended without reservation while a 4-star game is recommended with some reservation, etc. (i.e. a 1-star game isn't necessarily "worse" than a 2-star game, they simply fall into different categories in terms of recommendations).
@Inkerman said:
The number of stars do not translate into percentages, I believe the translation is as follows; 5/5 - PLAY THIS GAME!!!! 4/5 - This game is very good, we would recommend you play it. 3/5 - This game is good, but has some problems. 2/5 - Don't play this game unless you're a fan of the genre/series or can overlook the numerous problems. 1/5 - Do not play this game.
That is extremely dumb then, as games like Catherine got 2 stars. It makes no sense at all if you look at it like that.
@Alex said:
If this is being suggested because it somehow applies to Metascores or whatever other similar horseshit, no. Just no. We absolutely, positively do not give a rank shit about how our scores are portrayed on Metacritic.
It's also worth noting that Jeff has mentioned in the past that he thinks the Giant Bomb review scale lines up pretty well with Metacritic. 5/5 and 4/5, or 100 and 80 are green which suggests it's a good game that the reviewer recommends. 3/5 is 60 and yellow, which suggests that it's somewhat middling and has some issues that are hard to overlook, but can still be enjoyable. 2/5 and 1/5 or 40 and 20 are red, which suggests the reviewer thinks they are bad games that probably should be avoided. The way Metacritic presents the scores lines up pretty well with how those scores are meant to be taken in their original context.
@delt31: I find your premise entirely false. 4/5 or 5/5 makes perfect sense. 8,9,10 makes less sense, and actually pigeon holes more games into 9's, making the 5 point scale better at illustrating the reviewers feelings overall of the game. The less ambiguous these scores are the better, I always wondered why PC Gamer had a 100 point scale. WTF is the difference in a 88 and an 86 game? Whats the difference in a 6 and a 7? What do review "scores" even mater? I'm not comparing aggregate averages to make my purchasing decisions. If you're not complaining for the sake of metacritic, whats the rub, mario 3d got 4 stars? Seems pretty good to me and tells me "Hey, if you dig mario games, this is a pretty good one". What more does a review need to do?
Please Log In to post.
Log in to comment