next »

[ஒழுக்கின்மை]

it's probably an exaggeration to say most scores are bought. but it's not an exaggeration to say that most mainstream game journalists get bribes. it's not as simple as 'give me a 9 and you get this free hotel stay in hawaii' it's more like 'here's a free hotel stay in hawaii, thanks for reviewing our game chum'.

[Derek]

Number scores are useful in certain cases.  For example, on TIGdb you can sort by score, whereas "bottom lines" would be useless in that capacity.  What the sorting does is it gives people a place to start looking, starting with what the community deems to be the best indie games.  (Note: It's also easy to look at the very "worst" games.)  Of course, we also throw a bunch of random games on the front page, too.

Basically, any extra way you can categorize games makes it easier for people to get into them, in my opinion.  Are scores pretty reductionist?  Yeah, totally!  But they're also useful and if they're paired with reviews and other information I think it works out well.

Besides, the only way you can really know a game is to play it - everything else is just interesting fluff.

[ஒழுக்கின்மை]

i think the factors more important than official videogame reviews to me are demos, friend recommendations, and user reviews (the kind you read on amazon.com) -- i trust them much more than, say, metacritic score. but if there are no user reviews of a game, and nobody i know has played it, and there's no demo, then review scores are the best you can go by.

[TheDustin]

I think the biggest factor for scores is who is doling them out. In TIGdb it works, but for Play This Thing it's under-utilized; we only get a handful of people who score the games so they aren't exactly useful. I drifted away from mainstream gaming journalism when I drifted away from mainstream gaming, but I noticed on sites like Game Informer the "guest reviews" with scores of 10/10 and 0/10 have that meaningless "epic/fail" duality.

I do think a five star system is the most elegant though (I remember when Nintendo Power switched to that when I was about 9 and it made things easier on me), and in a perfect world we'd have both a one-sentence explanation and a score for the reading-challenged. 

[jwk5]

You could with Ebert's 4 star reviewing as it is pretty understandable (0-Very Poor, 1-Poor, 2-Average, 3-Good, 4-Very Good).

[JackieJay]

Scores can be really useful, they pretty much sum up the whole review into a single number.
I agree that 100% ratings are never accurate, I'm all for 5 stars or just a "passable","You must have it" and "complete shit" kind of rating.

Reviews and scores are a great way to decide if a game is worth buying or not, and indie games are no exception. Of course there would be little point to review a freeware game other than for the entertaining factor, but I was specifically talking about paid indie games.

and to paul eres who said the scores given by the mainstream game journalists are biased most of the time, well it's indie gaming we're talking about, I have my doubts that most indie developers would have enough money to pay the reviewer some holidays in the Caribbean. In addition, if the score is bought, usually so is the review. I can't imagine a reviewer saying the game is completely broken and then giving it a score of 90%, that would be just too obvious.
IMO, user reviews aren't any better. Like TheDustin said, they will either say the game is a complete piece of shit or they will say it's best thing ever made. Or will just lash at the reviewer because he didn't gave the score the readers wanted to see. If anything, I'd rather go for reviews written by people with a bit more experience and avoid reviews written by 14 year old kids who barely know how to write.

[Christian Knudsen]

I'm really hypocritical when it comes to numeric scores in reviews. On the one hand, I believe it causes readers to base their decision-making on a one-sided numeric value, whereas the reviews themselves will often be more nuanced. On the other hand, I always look at the score first when reading reviews (and sometimes just skip the review completely).

[Radix]

If they didn't show the score (i.e., they're hiding it from you), then you have to read through what they've said and guess as to what score they would have given it.

That is such a strange thing to think.

[Widget]

My (very personal) opinion is that a "score" is roughly as useful as the number of people who have contributed to the final reckoning. A 5-star vote system with, say, 1500 votes is a decent representation of a game's popularity (really not, necessarily, quality.. though it could correlate). A single person's "score" is only a value to slap onto their opinion. Whether it's totally honest or influenced by "presents", it's still just a single figure.

[C.A. Silbereisen]

People give scores to show if they liked a game or not. If they didn't show the score (i.e., they're hiding it from you), then you have to read through what they've said and guess as to what score they would have given it.

And what's so bad about that? I don't think it's that important whether the reviewer enjoyed the game, it's about whether you, the potential player, think you could enjoy it from the description he/she gives. One man's trash is another man's treasure and so forth.

Also, for most games I play, I have no idea what I'd rate them out of ten, and I guess game reviewers often face the same problem. There quite a lot of reviews, both in magazines and on the internet, where there's a massive discrepancy between the score and the "tone" of the actual writing.

[mjau]

Scores can be really useful, they pretty much sum up the whole review into a single number.

I don't think all reviews can be summed up into a single number.  Games are complex things, and boiling all that complexity down to a single number completely ignores all the little nuances and conflicting elements that make up whether or not you the reviewer thinks a game is good.  Like, say, a game has great combat but the story is complete tripe, or it has fantastic atmosphere but nonsensical puzzles, or even that the game is perfect in every way except audio, which is the worst thing you've ever heard.  What score would you give something like that?  Some just score graphics, music, etc individually and take the average of that, but this is a terrible way to do it since it completely ignores the relative importance of these things in the particular game being scored (eg an IF doesn't need good graphics), and those categories have complexities of their own in any case.  So if you want to do it "properly", it depends on the game and how important you think the good elements are vs the bad ones both specifically and generally, right?  But the answer to this is very individual and subjective and not always obvious beyond maybe a gut feeling.  That single number or amount of stars or whatever is almost meaningless without context.

I can't imagine a reviewer saying the game is completely broken and then giving it a score of 90%, that would be just too obvious.

Maybe not 90%, but i've seen this with 75%.  Makes you wonder what happened to the other three quarters of the scale.

[simoniker]

Also, please don't reference VGMWatch (at least, not the post-Kyle Orland version) as 'proof' that game journalists are all on the make, it's got crazy, EXTREMELY suspect sock puppet-y masters, as I discussed on GameSetWatch a couple of years back:

http://www.gamesetwatch.com/2007/10/vgmwatch_who_watches_the_watch.php

[Falmil]

Many people will have different preferences in games to the point where a reviewer's 9.0 score may be a player's 7.0 score. This is why reviews are meant to be read instead of just taking the score number. Fan-boyism doesn't help either where the player is offended that the reviewer dare give one of the games they've been looking forward to a low score. A lot of (bad) players simply want review scores that match their own opinions and will say the reviewer is an idiot if they disagree.
As it often turns out to be, the problem is on both sides.

[Craig Stern]

I decided to use scores in IndieRPGs.com reviews because they provide a reference point that many people find useful. Also, if you actually use the full 10-point scale, I see no reason why a 5-star system should be any better than a 10-point one.

Further reading: Jim Sterling's explanation for why Destructoid gives games scores,
http://www.destructoid.com/why-do-we-use-review-scores--157147.phtml

[JackieJay]

I don't think all reviews can be summed up into a single number.  Games are complex things, and boiling all that complexity down to a single number completely ignores all the little nuances and conflicting elements that make up whether or not you the reviewer thinks a game is good.  Like, say, a game has great combat but the story is complete tripe, or it has fantastic atmosphere but nonsensical puzzles, or even that the game is perfect in every way except audio, which is the worst thing you've ever heard.  What score would you give something like that?  Some just score graphics, music, etc individually and take the average of that, but this is a terrible way to do it since it completely ignores the relative importance of these things in the particular game being scored (eg an IF doesn't need good graphics), and those categories have complexities of their own in any case.  So if you want to do it "properly", it depends on the game and how important you think the good elements are vs the bad ones both specifically and generally, right?  But the answer to this is very individual and subjective and not always obvious beyond maybe a gut feeling.  That single number or amount of stars or whatever is almost meaningless without context.

Of course, the score isn't everything or else nobody would write reviews. What I exactly meant was that the score sums up the opinion of the reviewer. Obviously, just knowing he likes the game isn't enough for the reader to know whether or not the game is worth buying.  That's what the review is for.

Maybe not 90%, but i've seen this with 75%.  Makes you wonder what happened to the other three quarters of the scale.

Well in that case that's acceptable imo. Reviewers often exaggerate and label a game as completely broken just because it crashes every once in a while, which imo isn't enough to give it anything below 80 if other than that the game is great.

[Zaphos]

Also, if you actually use the full 10-point scale, I see no reason why a 5-star system should be any better than a 10-point one.

The full 10-point scale worse because everyone assumes you're using the more standard compressed 10-point scale and will misunderstand your ratings, and also in the comments section for every review with a less-than-perfect score you will get some angry comment that makes you feel you have to explain your rating system again.

[Christian Knudsen]

Yeah, I feel a 5 or 6 point score is better. With 10 points, there always seems to be the issue with some people feeling that anything below 7 is below "average", while other people seem to feel that below 5 is below "average" (this always seems to rile people up at GameTrailers). With 5 or 6, I think most people agree that anything below 3 is below "average" and should be avoided. With 5 or 6 points, there's also the possibility of adding a single adjective to the score, you know "poor, average, good, very good", whereas that's not really an option with the 10 point scores -- unless you want to employ "very, very, awesomely excellent!". 

[ஒழுக்கின்மை]

Also, please don't reference VGMWatch (at least, not the post-Kyle Orland version) as 'proof' that game journalists are all on the make, it's got crazy, EXTREMELY suspect sock puppet-y masters, as I discussed on GameSetWatch a couple of years back:

http://www.gamesetwatch.com/2007/10/vgmwatch_who_watches_the_watch.php

forgive me if i'm wrong, but nothing you wrote there is saying that they got anything blatantly factually wrong -- you're just basically saying there that they exaggerate and twist things to make it seem like a problem when there's not. which can be said of absolutely any watchdog site. plus, considering you're the editor in chief at gamasutra (or something along those lines?) of course you wouldn't like sites that criticize game journalism.

in any case, the site i linked to isn't important, it's just the best place for a collection of actual cases of bribery, and probably the most complete such collection, and i was asked to point to such cases -- so it's still the best answer to that question, whatever other problems it may have.

it's not where i learned of most of those cases. i've just heard of them over the years in dozens of different places. many of which places i no longer remember. for instance, i remember that someone once called edge demanding that an 80 become a 90, or they'd pull advertising money. but i don't remember *where* i heard that anymore, because it was so long ago. it may have been in egm, or game informer, or gamespy, or some online article on reddit, i don't know.

[Aquin]

Look, I know for a fact why scores exist.  It's not for discussion.  It's for purchasing.

Every single customer I had (that wasn't a mother) knew the score on any given game.  "Yeah that game was actually pretty good," I'd say.  "Oh really?  Because it only got a 7 out of 10 on GameSpot."

Like, how the fuck am I supposed to argue with that?  Trust me, those scores are used *heavily* by many consumers.  Some of them didn't care, but most of them bought games blindly based on score.

Okay, but we need scoring.  You know what system I like?  IGN's comic review system.  I like neither their critics nor their conclusions, but I can't argue with the usable system.

Buy It
Try It
Forget It

Because people reading a review, that's usually all they care about.  They skip to the bottom just to see the recommendation.  The only people arguing the finer points of your review arguments?  Other critics (or forum-goers who are amateur critics.)

[JackieJay]

Exactly.
Some people might like to piss over most reviewers's heads but the truth is they follow them almost religiously.
Just reading a review doesn't tell you exactly how much the reviewer loved/hated the game. That's what the score is for.