Support NeoGAF

[ViewtifulJC]

Because a personal essay is the closest you can get to communicating a game aside from playing the game itself. It's obviously a lossy compression of information, but in the end it's the best we've got.

And it's still more useful than a single number, a number for which correlates to a table of extrapolations nobody can seem to agree on.

Its not an either/or thing here. There's text AND a number, they go hand and hand, they support each other(or at least that's the idea). Its not like Destructoid said "Infamous 2 review inside" and you click the link and only find a number out of 10 there.

 

[fedexpeon]

Review scores are based on "Asian Parents Scaling".

A+: 10, you are doing great!
A: 9-10, you are doing good.
B:8-8.9, you are doing average.
C: 1-7.9, get the hell out of my house.

 

[Aquamarine]

Time to create an addblock like program that hides the scores for reviews.

What about review sites that only show you the score after you answer five questions about the content embedded within the review?

 

[AndyD]

Marketers know bigger numbers = better in the human mind. So 5 on a 10 point scale will never be average or satisfyingly normal. It will be completely terrible because it's so far below 10. For some people even 8 is a bitter disappointment.

What dev was it that missed their bonus for getting 84 not 85 on metacritic?

 

[Stumpokapow]

Because people scale out of 10. It just inflates everything needlessly leaving the "pit of death" below 6 which is hard to actually understand. What's the difference between any of those? Nobody knows what really pushes people to give something a 2 as opposed to a 4, they're just two bad scores.

5-point scale is all people could ever need:

Garbage, nobody should suffer through this
Bad, but maybe you're a fan so go ahead
Passable, worth a shot if you're interested
Great, you should look into it even if you're not interested
InstaClassic, everyone should play it

Simple, and every tier has an actual meaning. Unfortunately for publishers though it would be dragging down their metacritics a lot (no more 9/10s)

Whether there's granularity in the scale or not, there's granularity in the review. This is true of any scale. In a yes/no system, there are borderline yes and borderline no reviews as well as decisive yes and decisive nos. In a 10 point system, there's products that seem better than the rubric for an <x> but not as good as an <x+1>. Eventually you decide you're hair splitting and move on, but the initial reaction is always going to be "wellll, it doesn't quite evenly fit into the boxes we have". It's something every review struggles with, including things like markers in social sciences and humanities courses, or judges in scored competitive events.

 

[Ravidrath]

The 6/7 thing was a problem LONG before Metacritic existed.

It's because people map it to grade school percentages.

Metacritic introduced all the pressure PR people put on reviewers, how they game the system, etc. to try and maximize their score.


The best review system would be Red / Yellow / Green, or Not Recommended / YMMV / Recommend. It's how people actually process things, and they have to read the text to determine if they actually want it.

But the industry would never allow it, because they've built everything on Metacritic scores now.

 

[Phire Phox]

The gamers. If you think a 7 or 8 is a bad score and you won't buy a game, that's your problem.

 

[scoobidoo112]

I think it definitely results in situations where a lot of people don't click on the review and just react to the score. It's not really that much of a problem here because the reasonable people will actually read the goddamn review, but ideally, it would be nice to have a be a policy in place where you just post a link to the review to promote actually reading it.

Like this:

IGN Dark Souls 2 Review
GameSpot Dark Souls 2 Review


Instead of this:
*image*

Wow, those review scores are ridiculous. Have the reviewers even played DS2? Don't get me wrong, it's not a "bad" game at all, but with the visual downgrades a month before release, the messed up hitboxes, the 180 degrees mid-attack spinning enemies and overall downgrade as opposed to the first Dark Souls, how is that game getting 9/10 or 10/10?

 

[pokeystaples]

I think it definitely results in situations where a lot of people don't click on the review and just react to the score. It's not really that much of a problem here because the reasonable people will actually read the goddamn review, but ideally, it would be nice to have a be a policy in place where you just post a link to the review to promote actually reading it.

Aquamarine is on point per usual.

 

[Danny Dudekisser]

I blame it on Gamefan and their ridiculously inflated ratings in the mid to late '90s. Didn't Final Fantasy VII get like... a "100+" or something ridiculous like that?

 

[Nymphae]

I don't understand why review score uses decimal point.

Like 8.8, but not round it to a 9.0.
I know IGN believes it does matter in determine what is great and what isn't.
But .2? WTF is the difference?

Using .5 I can understand though, it is better than the lesser whole # but not better than the other greater whole #. It is just in between.

But a .1-.4, .6-.9 range? They are tacking in points from a subjective viewpoint.
Yeah that game definitely deserves that extra .1 point for having a MC with brown eyes. /s

The pinnacle of this shit, to me, is Tommy Tallarico's reviewing games on "Reviews on the Run" back in the late 90's & early 2000's. He would just through arbitrary decimals around like they actually meant something. 8.7, 9.3, 7.1....come on dude, pick a goddamn integer.

 

[zackmorris]

Honestly its so easy to set up a half decent website. Call yourself professional and then throw a wonky score in as clickbait and you have half a reason.

What does this have to do with having a skewed scale? Doesn't matter how many "wonky" scores a game gets, he's asking why a 7 is considered bad.

 

[Ravidrath]

I blame it on Gamefan and their ridiculously inflated ratings in the mid to late '90s. Didn't Final Fantasy VII get like... a "100+" or something ridiculous like that?

Haha, GameFan gave a number of 100+'s. Crash Bandicoot got one, I remember.

They also gave Trevor McFur a really high score, because Dave was literally tripping on acid when he played it.

 

[hardcastlemccormick]

The only way in which a score is more reductionist than a narrative is if you believe the score and the narrative diverge. I do not typically see this in reviews; typically the reviewer feels that his or her final score represents his or her opinion. Even reviews with meaningless subcategories tend to have very clustered scores (IE games that are viewed as bad thus end up having bad graphics scores, bad music scores, bad gameplay scores rather than scattershot category scores and a negative conclusion).

Can you provide examples of cases where the score provides an impression that the text does not, to the point of misleading readers? Note that IGN's 2.0 for Deadly Premonition is a totally shitty review, but it's not a shitty review because they gave it a 2.0, the 2.0 matches the review text well. It's a shitty review because it's a shallow, surface-level examination of the game that misses the forest for the trees and is myopic in that it failed to recognize the extremely divisive and complicated reception the game eventually got.

I believe you are misunderstanding my point. I am not talking about a score that is paired with a review, but rather a score in isolation.

If I play Deadly Premonition and give it a score of 3, sans any justification or text explaining myself, then that is an effectively useless reduction, which I categorized as "offensive," a term I still stand by.

If I play Deadly Premonition and write a 20 word essay, I've compressed the experience of the game in a better way than a single number, but it's still fairly light on details. The more words added (assuming each word adds the same amount of useful amount of information) then the game will be better and better communicated.

Now, if your question is, "what do you make of scores and reviews paired together" then I would argue that they're fairly useless as well, unless there are people who WANT something so incredibly reductionist that they can consume the point in a single thought. Otherwise, it's the same communication, just worse.

Have to disagree with you there. Just because something took a lot of man hours doesn't mean it is inherently worth any respect. The amount of code or the number of people that built something is totally irrelevant to a consumer's impressions and final evaluation of a product.

Absolutely nothing wrong with a binary or 5 point scale. I think most people can easily distinguish between "worth it/not worth it" and "hate/dislike/mediocre/like/love" and, ultimately, it is how we categorize most things that we experience.

It's not a matter of "worth," it's a matter of information density. I can compress the experience of playing a shitty game in a poor way, much like you can poorly compress a book into a zip file and lose substantial information to the point of uselessness.

 

[MHWilliams]

No "Maybe." Our scale is binary.

Even if it's binary, the point stands: 1-5 = "No", 6-10 = "Yes".

Reviews are subjective and ultimately we're distilling a creative work down to a product review. Which makes sense because they're frequently $60 products and some people just want help making their purchasing decision.

Stump is right.

Whether there's granularity in the scale or not, there's granularity in the review. This is true of any scale. In a yes/no system, there are borderline yes and borderline no reviews as well as decisive yes and decisive nos. In a 10 point system, there's products that seem better than the rubric for an <x> but not as good as an <x+1>. Eventually you decide you're hair splitting and move on, but the initial reaction is always going to be "wellll, it doesn't quite evenly fit into the boxes we have". It's something every review struggles with, including things like markers in social sciences and humanities courses, or judges in scored competitive events.

Readers care about scores for various reasons, otherwise review threads on here or Reddit wouldn't pop up immediately. Fix that and review scores would go away accordingly.

 

[danwarb]

People who review games/movies/music for popular outlets are not the most sophisticated of journalists. It's a safe way to proceed. There's often outcry from emotional fans if a game from a company they're invested in scores too low, like 8.5. It's what the people want, not what's best for them.


Don't bother with them, find something better and make that popular.

 

[Nydius]

Its not an either/or thing here. There's text AND a number, they go hand and hand, they support each other(or at least that's the idea). Its not like Destructoid said "Infamous 2 review inside" and you click the link and only find a number out of 10 there.

Except they often don't go hand in hand. How many times have you read a review that seemed more negative than the final score or vice versa? Happens all the time. Text reviews that sound like a game has lots of technical flaws but gets a 9 or a text review that praises the game and points out flaws but then gives it a high 9 or 10. The 10 scale rarely goes hand in hand with the text review.

I have to stick up for Jason here: Kotaku's "should you play: Yes/No" is probably the best current review scheme. It doesn't detract from the text review and offers a binary opinion by the author. For the longest time, movie reviews used the "thumbs up/thumbs down" review benchmark made popular by Siskel & Ebert and those reviews turned out just fine. Why do games need a ridiculously large 10/100 point scale that can be manipulated through decimal points? Text review plus a simple "In my opinion, this is worth playing" is more than sufficient.

Edit:

Even if it's binary, the point stands: 1-5 = "No", 6-10 = "Yes".

You're really reaching here. There's a vast difference between 1-5/No and 6-10/Yes. The former can be manipulated through various methods in order to obfuscate points or make ridiculous arguments (like IGN's arguing that their 8.9 score wasn't really a 9.0) while the latter is a straight binary opinion that doesn't get shit tacked onto it like ".2 for [insert inane reason here]".

 

[flatearthpandas]

I believe you are misunderstanding my point. I am not talking about a score that is paired with a review, but rather a score in isolation.

If I play Deadly Premonition and give it a score of 3, sans any justification or text explaining myself, then that is an effectively useless reduction, which I categorized as "offensive," a term I still stand by.

If I play Deadly Premonition and write a 20 word essay, I've compressed the experience of the game less than a single number, but it's still fairly light on details. The more words added (assuming each word adds a useful amount of information) then the game will be better and better communicated.

Now, if your question is, "what do you make of scores and reviews paired together" then I would argue that they're fairly useless as well, unless there are people who WANT something so incredibly reductionist that they can consume the point in a single thought. Otherwise, it's the same communication, just worse.

Who ever suggested these scores were in isolation in the first place?

More importantly, offensive to whom? What about all of these lines of code or hundreds of people makes the game worth more than a single digit reaction?

I think reviews include text not because the score itself is offensive to most readers but to justify and explain the final impressions, i.e., the number score (in most cases) so that the reviewer, not the game, can be understood.

Didn't notice the edit:

It's not a matter of "worth," it's a matter of information density. I can compress the experience of playing a shitty game in a poor way, much like you can poorly compress a book into a zip file and lose substantial information to the point of uselessness.

I think this raises the question of why people are reading and writing reviews in the first place. I'll insist that the hundreds of people and number of lines of code are irrelevant in any case.

I don't think reviews are written or read in order to experience even a simulacrum of the full experience of the game. The point of the review is to capture and share impressions of the full experience, which requires the experience of the review to be reduced from the game. I believe the shortest way in which we can meaningfully share these impressions is not "offensive" but instead "ideal."

Of course, I'm not saying that a standalone number is the shortest way to meaningfully share these impressions. Not at this point, at least.

 

[sn00zer]

Games pretty much stopped being terrible.....A game deserving of a 1-4 score is a rare anomaly these days
Even bad games these days have at least a few redeeming characteristics. Its too expensive to make a truly terrible game anymore.

 

[PBY]

Games pretty much stopped being terrible.....A game deserving of a 1-4 score is a rare anomaly these days
Even bad games these days have at least a few redeeming characteristics. Its too expensive to make a truly terrible game anymore.

We should stop reviewing games as products. There have been perfectly fine games that have left me cold and bored. If I reviewed them, is give them 1-2 stars.

 

[Aquamarine]

People who review games/movies/music for popular outlets are not the most sophisticated of journalists. It's a safe way to proceed. There's often outcry from emotional fans if a game from a company they're invested in scores too low, like 8.5. It's what the people want, not what's best for them.


Don't bother with them, find something better and make that popular.

That's such a broad generalisation. How do you define "sophistication"? And what's a "popular outlet"? Is Destructoid a popular outlet? Is Total Biscuit's YouTube channel a popular outlet?

If I were to rebut this argument, I'd say that there is a healthy mix of talented and mediocre journalists at any level of journalism. That's the inherent nature of professions when the value of the inputs and the outputs are subjective.

I hope you don't mix up brevity of a review with "lack of sophistication."


And what's this about a conspiracy? Game reviewer choose high scores because they don't want to upset people? Give me a break.

 

[Zimbardo]

my scale kinda goes something like this ...

6 is average/so-so
7 is good
8 is very good
9 is excellent
10 is amazing/outstanding

5 and below are varying degrees of shit.

 

[hardcastlemccormick]

Who ever suggested these scores were in isolation in the first place?

I did. I'm talking purely theoretically, in terms of taking information and compression or reducing that down into smaller forms.

EDIT: Note that the OP is not talking about how review text is broken, just review scores. This is where my train of thought is heading.

More importantly, offensive to whom? What about all of these lines of code or hundreds of people makes the game worth more than a single digit reaction?

Offensive to me.

Again, it's not about "worth." I said this once and I will not say it again.

A string of 1000 random numbers has no worth, but reducing it down to a single digit is useless and offensive. This is my point.

I think reviews include text not because the score itself is offensive to most readers but to justify and explain the final impressions, i.e., the number score (in most cases) so that the reviewer, not the game, can be understood.

If a score needs text to justify it, then the score is not sufficient, whereas text can exist without a score alone. Therefore, the text is superior in all aspects.

 

[Dr Dogg]

I get the feeling that a fair few people use reviews to validate their purchase and not inform them of a potential one. I think a lot of us are already sick to death of the "preordered canceled" posts that keep on cropping up.

I've got my go to set of peeps on GAF who I value their opinions more than people who review professionally, not because we share any common interests but because I know what they like. Something they might hate I might really like and vice versa. Some of them are way better at articulating their experience with a game better and they are not under any obligation to get an impression out for a set date any more than any of us are to make a day one purchase.

 

[Freeman]

Games cost $60, that is the root of the problem to me.

I'll gladly pay to watch a 6 out 10 movie, who knows, I might like it. I use the aggregate sites to get an idea if isn't a total disaster(when I'm unsure), but just by knowing who is involved you can guess with a very accurate rate if its worth to watch. Worse case scenario I regret spending my $4 and 2 hours of spare time.

When it comes to games, it either needs to be a great game (GotY candidate) or cheap (and still pretty good), I won't be happy to pay $60 for mediocre or even above average games.

The average game sucks, I'd rather not waste 20hours on something I don't like very much. Too bad lengthy game completion time is still seen as a good thing, to me unless the game is excellent or make great use of its time, its a negative. I use reviews just to see if there is some major technical problem, otherwise its useless to me. If I'm unsure if the game is for me, I'll just watch a little bit of it and I usually can tell if its worth it for me.

 

[Trigonometrize.]

Ignore scores, read words.

 

[HoodWinked]

It's based on grading in school. 50% or less is considered failing. So me terrible game that fails to work would be a 5 or under. A "B " level game is a 7-7.9. An "A" game is 80-89 and an "A+" game is 90-100. It's just the way we were graded in school.

where do you go to school?

 

[Gestault]

Stumpokapow is single-handedly laying waste to a lot of undeveloped concerns in this thread. It's quite refreshing.

 

[Hyun Sai]

Because it's how grades also work in real life ?

 

[Holykael1]

Reviews are worthless anyway, atleast for me. Once you know what you like you can make pretty accurate educated guesses on how much you will enjoy a certain game by drawing parallels to other experiences you had and by using the information available to you of that product(looking at assets, previews and such).

Had I followed the whole review thing I would have missed out on incredible games.. Only you know yourself to the fullest extent, I understand wanting to feel safe when buying something but I'd rather have 1000 disappointments than miss out on a game that could potentially "change my life".

So far my method has been working out pretty great, I never bet on something that I didn't end up enjoying even if not meeting expectations and I have put my faith in other games that were trashed by other people only to come out with a big positive surprise.

Since gaming is an inherently subjective experience, just like listening to music, reading a book or watching a movie, it works fantastically because everyone is different and appreciates or has varying degrees of tolerance towards certain aspects.

I only take the word of people as information, being aware that something exists and learn about it's characteristics, besides that I'd rather judge things for myself and that's that.

Edit: Opinions are more valuable to me when I know the person in question moderately well, there are a lot of variables to take into consideration..

 

[Aquamarine]

Stumpokapow is single-handedly laying waste to a lot of undeveloped concerns in this thread. It's quite refreshing.

Well he is a Forum Administrator here at GAF...I'd expect someone like that to present intelligent arguments given the generally-high caliber of mods. ;-)

 

[Shorty11857]

In all honesty a lot of it depends on what you class a 5 as; is 5 average or OK? Most AAA games are not bad, I honestly can't remember the last time I thought a big budget game was bad. I mean that's not to say they're all good but for the most part they'll be competent at the very least which easily gets them to a 5

Honestly I really think games just need 5 ratings, Bad, Okay, Good, Great, Amazing.

 

[JayhawkerStL]

Every gamer that mentions meteoritic in checking reviews is part of the problem.

 

[Kai Dracon]

Haha, GameFan gave a number of 100+'s. Crash Bandicoot got one, I remember.

They also gave Trevor McFur a really high score, because Dave was literally tripping on acid when he played it.

In one editorial I believe Dave said that Gamefan gave out such great numbers because they just never reviewed bad games.

 

[dimb]

Read the review, ignore the score. Magically unbroken.

I feel like almost every review I read is still shoddily written and incapable of identifying the strengths and weaknesses of a title. There are huge components of games that can go completely missing from almost the entirety of games criticism. Even just recently the latest Kirby game has the most protracted and lazy final world that amounts to little more than all of the bosses being retread without any new components to them. Maybe people like this, maybe they don't, but it's a pretty big element to a game that's already only six hours long to begin with, and I didn't see this mentioned by any of the reviews I scanned.

 

[Nickle]

If you get a 50% on a test in school you are not average. That's how many reviewers use the 10 point scale.

 

[Ravidrath]

In one editorial I believe Dave said that Gamefan gave out such great numbers because they just never reviewed bad games.

Well, they also actively massaged scores of games from publishers that advertised in that issue.

Editors didn't give the final scores - they'd make recommendations, and then Dave would assign the final score based on ads that month.

 

[Alebrije]

Because reviews are now part of most videogames marketing , you just need to see them as adds.

Long boring adds btw.

 

[Linearskillz]

Guess I'm part of the problem because I like scored reviews. What's important is that you interpret the score relative to other scores instead of at face value. It's true that the low end is rarely used but I don't think that's inherently a problem unless for some reason you're convinced that 50% should be the average review score. If you really wanted to you could bell curve everything to average 50% but that wouldn't really change the interpretation.

You could argue all day about how it's reductionist to represent the quality of a game with a number, but myself (and probably most people) don't want to put that much time and effort into every single review to understand whether or not a game is good. I see a game and ask "Is this worth buying?" and I want to be able to answer this question quickly and efficiently. I'll read/watch reviews from my favourite publications but otherwise glance over the scores from everywhere else, and I honestly don't think that's all that bad. Even if all reviewers stopped giving scores simultaneously I would bet money that a service would pop up that distilled the text into short form numbers (Rotten Tomatoes anyone?).

With regard to review threads I like them because they often have review listing long before Metacritic or Game Rankings does. This also means they become pretty useless to me once Metacritic catches up.

 

[MHWilliams]

You're really reaching here. There's a vast difference between 1-5/No and 6-10/Yes. The former can be manipulated through various methods in order to obfuscate points or make ridiculous arguments (like IGN's arguing that their 8.9 score wasn't really a 9.0) while the latter is a straight binary opinion that doesn't get shit tacked onto it like ".2 for [insert inane reason here]".

What's the dividing line between the Yes and No? Which feature pushed you for a Yes to a No? Is it the feature A's implementation or the story? I think the story is fine,so that'd be a "yes" for me.

The .2 or .3 you mention is still just a subjective construct created by the reviewer.

Either way, you're boiling down the (hopefully) nuanced review into a single pithy recommendation. Arguing whether Yes/No, 3 star, 5 stars, or 10 stars, or 100 points is better or not doesn't change that. If you're against review scores, then really be against review scores. Text-only reviews that you actually have to read the whole thing and process.

 

[Kai Dracon]

Well, they also actively massaged scores of games from publishers that advertised in that issue.

Editors didn't give the final scores - they'd make recommendations, and then Dave would assign the final score based on ads that month.

Ha ha - I swear hearing GameFan anecdotes as an adult is like discovering Santa Claus isn't real over and over again.

 

[rjcc]

If you think the important part of the review is the score, you're the one who is broken.

 

[Cymbal Head]

Has anyone ever stopped to consider the proposition that all games are better than average?

Just sayin'

 

[hardcastlemccormick]

Didn't notice the edit:
I think this raises the question of why people are reading and writing reviews in the first place. I'll insist that the hundreds of people and number of lines of code are irrelevant in any case.

I don't think reviews are written or read in order to experience even a simulacrum of the full experience of the game. The point of the review is to capture and share impressions of the full experience, which requires the experience of the review to be reduced from the game. I believe the shortest way in which we can meaningfully share these impressions is not "offensive" but instead "ideal."

Of course, I'm not saying that a standalone number is the shortest way to meaningfully share these impressions. Not at this point, at least.

The only reason I mentioned the hundreds of people working on a game was to emphasize how much information is packed into even the shittiest game, not that hundreds of people make a game "worth" more.

We agree that reviews are a necessary way to communicate a game to those who have not played it. I gained a good-enough understanding by reading Wolfenstein's recent reviews, and in the end I will likely buy it because what they describe is something I would enjoy quite a bit from my tastes in FPS games.

Where I think I may not be expressing myself correctly is in your sentence:

I believe the shortest way in which we can meaningfully share these impressions is not "offensive" but instead "ideal."

I agree with this! The issue is that review scores cannot meaningfully share impressions. Review text can, if it's sufficiently long. But if the text is already doing a far, far better job than the score is, then the score is unnecessary.

 

[molnizzle]

When I went to school, which wasn't really all that long ago, anything below a 70 was considered failing. When did it get changed to 50 or below was failing?

Yep.

A = 90-100
B = 80-89
C = 70-79
F = 0-69

That's pretty much how I interpret game reviews as well.

 

[aeolist]

Has anyone ever stopped to consider the proposition that all games are better than average?

Just sayin'

ur blowing my mind dude

 

[mcz117chief]

The games got better. At least I feel like it. You really can't strike down any game that got 9s lately, or can you ?

 

[mr teaspoon]

The gamers. If you think a 7 or 8 is a bad score and you won't buy a game, that's your problem.

Exactly.

 

[spekkeh]

Some short term memory here. 7-10 scores predate metacritic by some time. It happened because in the olden days magazines would base their rating on 'objective' measures such as graphics, sound, longevity. Basically games got a review as if they were hardware appliances (hence the name review instead of critique really). The overall score was an average of these numbers.

With increasing computing power and dev sizes, came increasing graphical fidelity, sound composition and game longevity. Game iteration 2 was 'objectively better' than game iteration 1, and thus got a higher score.

This was compounded by game franchises getting more iterations and kept on going somewhere halfway into the N64 period, when all new AAA games ended up scoring between 95-98.

It started normalizing afterwards.. slowly. This is why Twilight Princess getting a 88 was major major bomba. It was unheard of that major games would score below 95, think about all the time that had gone into that game.

But everything below an 8 still feels completely worthless.

 

[Bloodrage]

Your game got a 8 lol floooooop