In Defence Of Perfect Scores

You will see, occasionally, the complaint that a video game has been given a perfect score. 10/10. 5/5. All five stars. The complaint tends to be that there are too many of these perfect numbers. These scores must come, the thinking goes, from easily impressed reviewers who are as full of bias as they are breathless.

Some reviewers support this reaction by saying they will not give a perfect score. Nothing, they say, is perfect.

I give that line of thinking a 6 out of 10.

I'd like there to be more perfect scores.

Perfect scores do not indicate perfect video games, because there is no such thing as a perfect video game. No created work that is indisputably without flaw.

If we believed games could be perfect, what would be our exhibits? The eternal Pac-Man, though it might not have been as fun as Ms. Pac-Man? The ubiquitous Tetris, which has had soundtracks you may not like? What of the dominating Angry Birds, which pleases millions while still not fitting quite right into some mobile phone screens? How about the excellent, aged Legend of Zelda: The Ocarina of Time with its annoying fairy helper? Or the magnificent Super Mario 64 and its muddy graphics? Can we give the best praise to Metal Gear Solid, except when its controls betray it?

If we will spare the world's best video games the highest statistical praise, let's demote our praise of the Sistine Chapel for requiring the bending of one's neck to see it. And let's lament that Michelangelo didn't capture a Mona Lisa-quality smile on either God or man. The best movies, sculptures and plays I have ever seen have had flaws and inconveniences, as have the best books and songs.

If we believed games could be perfect, what would be our exhibits?


At Kotaku we "score" video games with a Yes/No/NotYet in response to the question "Should you play this game?" We do this as we endeavour to give the readers of our reviews a more natural response to a more natural question about the quality of a game. But I do not begrudge our peers who utilise taste mathematics to quantify the quality of what they have played. Many people like a number. Many people don't just want to know if a video game is fat or skinny with quality. They want to know what its quality weighs, assuming such a measurement can be ascertained.

Today, our friends at Polygon reviewed Diablo III, which may or may not be a most excellent game. We gave it a Yes. They gave it a 10/10 and yet their reviewer acknowledged the game was not perfect (highlight added by me):

Cue the response from a reader protesting that the number did not match the caveat. And cue subsequent readers who said those two things could co-exist.

Let us give more games the highest honours. Not because games need to be graded on a curve. Time already corrodes most games as advances in technology make many games tough, in retrospect, to enjoy. The best games already crumble from the withering impact of advances in design. Must great games also suffer a stinginess born of an impossible standard.

No video game is perfect. But so many of them are perfect 10s.

Top photo: rimira/Shutterstock


    I don't agree.

    I think a video game can be perfect. It's subjective, sure, but I think it's reasonable.

    I, fr example, couldn't cite a single flaw in Braid, nor Commander Keen, nor The Amazon Trail, nor Pong, nor Space Invaders, etc, etc, etc...

    Sure, you may not agree that those games deserve 10s, but they, to me, subjectively, are intrinsically perfect.

    Diablo III has far too many issues to be considered a 10, in my opinion.

      I think applying a number rating to a game is completely arbitrary at the best of times and that gaming is too connected to a person's personal experience for us to objectively call any game good or bad as long as it functions properly most of the time.

      I think games need to be rated with respect to the 'promise' that the game is essentially marketing itself as. This way you keep your scores relative to the genre, platform, studio, budget etc... that it was made for. If you don't then trying to compare games across these boundaries, or putting a reviewer who's used to a different area of games into something new to them will without a doubt skew their opinion.

      Also I think despite what the companies say, studios and publishers have far too much influence over game ratings websites. Maybe someone could come up with a game advertising company that conceals the website on which the advertising is put from the one posting the adds, meaning that the producer can't retaliate by pulling advertising dollars from a review site because they can't control who it goes to.

    I always liked the out of 100 thing N64 Gamer magazine did when I was a teenager, only games that ever got 'perfect' scores in that were Super Mario 64 and Perfect Dark.

    The problem is not with 10s/5 out of 5s/etc.

    The problem is with scores of 7 and below.

    A 4/10 game might actually be worth your time in some way, but sadly the only time reviewers make a big deal of them is when they are laughably bad/broken (that recent XBLA survival horror game for example).

    there has been many games in my long gaming history that id give a 10/10 or a 5/5, however if i was to break that down further they would end up as say a 98/100.

    Deus Ex:HR was a 10/10 in my books, however if i was to break it down id find flaws, so 98/100 would be more accurate.
    Batman arkham asylum/city both 10/10 however i did not like the controls for both so out of 100 both would get 95.

    Perfection cant be measured as a number, and is subjective to the content and the person experiencing it.
    the only game that i'd give a rated perfection 100/100 would have to be alex the kidd in miracle world, no other game in my long and horribly expensive gaming life has ever captivated me and kept me so entertained, but that in itself proves that theres no such thing as perfection in games, cos that game was riddled with problems lol

    I think the opposite way actually. Too much fanboy praise, not enough critical thinking.

    There have definitely been games I've played that justify a perfect score - Both Mario Galaxies, MGS1, Tony Hawk 3, Half Life, and so on ... however, each of these games have dull levels, or annoying glitches, or bits which just don't seem to fit. It doesn't mean they're any worse - it shows that they're "human", in a sense. Sometimes it's the flaws which accentuate how good the rest of the game is, and that's what makes it perfect.
    I've played platformers where the absolute pinnacle of gameplay are equal to the worst part in Mario Galaxy; that's what sets it apart, and makes it perfect.

    The problem isn't with giving an imperfect game an 10/10 the problem is that rating systems like this are poorly defined, even more poorly implemented and overall meaningless.

    There are a number of games I can think of which I would sing the praises of to the high heavens if I was to write a review. Not a single one of them I would try to give a numerical score to.

    Red Dead Redemption would rate highly for entirely different reasons to Journey. Giving them both a 10/10 only means that I think that they're really bloody good. If I was writing a review about them, you'd have known that long before getting to the arbitrary numerical summary.

    No the system is broken. Reviewers give average games (games they themselves label AVERAGE) a 7/10. 7/10 is not average. 7/10 is ABOVE average. therein lies the problem.

    If we could go back to giving an average game like MW3 a 5 or 6 and then an outstanding game like an ME2 or Bioshock an 8 we wouldn't have this problem. This confusion. No game is perfect, but if you want to give a 10 it should be reserved for not only a truly brilliant game, but a game that pushes boundaries, that tries things that haven't been tried before or often - and pulls them off - this is a game that could deserve a 10. But when Call of Duty: 38 More Guns is given a 10/10, it demeans the entire system.

    Has anyone of note ever reviewed the Sistine Chapel or Mona Lisa as being 10/10? Have they claimed them to be perfect? No. Thus your entire article falls apart.

    The only time a 10/10 is justified is when you only score in integers, and the sum of all of the flaws in the game amount to less than 0.5 points.

    However, the argument has rarely been that 'no game should ever be reviewed as perfect', but rather, the scoring system is imbalanced. It's the sliding scale of 7 to 11 out of 10.

    If a film gets 3/5 stars, it's a good movie. If it gets 4/5 stars, it's a must see. If a game gets less than 8/10 (more or less the same as 4/5), it's considered a failure.

    5/10 should be average. A good AAA release should be around 8/10

    I personally never give out perfect scores. But I'm fine with games that are really good and should deserve a 9 being given 10s by professional reviewers e.g. Skyrim, Portal. But not clearly flawed games that deserve 6s and 7s being given 10s. That will make me give it something like a 3 or 4 to counteract the 10 e.g. Dragon Age 2, SWTOR.

    Games can't be perfect?
    Two words.



    I dont think its a matter of too many games given perfect scores, I think its the fact that games only get scores between 7 to 10.
    Reviewers have given themselves no room to move on scores with such a narrow scoring field, so a 10 is too easily given.

    The problem the games industry has is that there is no standardised rating system. Movies almost always get rated out of 5 stars and people know that one critic's 5 star movie isn't another one's 5 star movie simply by viewing the rating and then reading the review itself to find out why one critic hates it while another one loves it. After reading enough reviews they get a sense of what each critic likes and doesn't like and then chooses the critic that reflects their tastes the closest. For example Margaret Pomeranz by far has the most similar taste in movies to my own so if she recommends a movie 999 times out of 1000 I enjoy it too.

    Make a standardised industry-wide 10 point system (1 to 10 in whole increments or a 5 star system with increments of halves) with points/stars allocated to games as each site/critic/reviewer sees fit. At least then the metacritic style sites are able to provide more consistent and statistically significant results though I'd prefer these to adopt a more Rotten Tomatoes type of rate where they offer a percentage of critics who like a game (ranked 5/10 or above) and give it a fresh/rotten or yes/no type of meta-rating.

    I prefer a review which break down each aspect of the game..
    There can never be a perfect 100% but that's fine, each person can read it and look at the score of what is important to them ie graphics or game play etc etc and decide whether they want to buy it.

    Some times reviewers get lost in their own article and score does not reflect the writing.
    + I think Kotaku should add a viewer rating poll on to the user reviews.

    If everyone just let go of the idea of scores being how good a game is and instead treated them as how much the author would recommend it, we'd have far less of a problem with review scores. Frankly, most games these days are competent and fun, which is why we have the four-point scale. It'd also eliminate some of the cliches of reviews, where reviewers try to explain that fans of a genre should check out a game in that genre. That's 3 stars. Done.

    I generally don't put that much effort into predetermining any purchase based on a stranger's opinion. I'm more hands on.

    Part of the problem is that the scale has become skewed. It's unusual to see a game get less than 5/10 in a review, and if it does, it usually means that it's absolute rubbish. To me 5-6/10 should represent an "average" game, not great, but not crap either. Good games get 7 or 8/10, excellent games get 9 or 10, terrible games get 1 or 2. Reviewers seem hesitant to give out <5 star reviews, and it seems they're starting to be hesitant about giving 10 star reviews, due to the reader backlash about "perfection". I'd like to see a more normalised scoring system, and more reviewers who are willing to be honest about how good or bad a game really is.

    I think most people just come up with an arbitrary idea of how things should reviewed then hold every reviewer to their ideal. There isn't even really an argument here, people will clumsily try to explain how the professional writer is wrong in their opinions but really, it doesn't matter. I read reviews to hear others' thoughts, not to have my own echoed in a different voice.

    The main problem I have is with consistency. There isn't any. Fine, a game can get 10/10, I don't care about that. But if you compared two games together, for instance, two games which haven't improved on any sort of formula, same gameplay mechanics, etc, you find that a lot of the time these two games will score radically differently. You'll also find that a lot of games nowadays will score very well despite having a number of obvious flaws which other games will score lowly for (Splinter Cell Conviction got chewed out for always online DRM whereas many reviewers have blatently ignored that with Diablo 3).

    Giving scores is just stupid in the first place. For so many different reasons.
    Reviewers should just talk about the good/bad points and the viewer can make up their own minds.

Join the discussion!

Trending Stories Right Now