The original Eurogamer scoring system (pre-May 2009)

Wondering how we pin a number on a game's worth? Here's our scoring criteria laid bare...

Scoring systems are, by and large, a black art. Anyone who has kept track of the scoring of games over the past 20 odd years will be aware of all manner of systems, none of which are ever going to be truly satisfactory to everyone. While some will argue to the death that percentages are the best and more exact means of judging a game's worth, do we really know the difference between a 75 and a 77 per cent game? Of course not. In such a subjective area as someone's opinion (which is what a review actually is), quibbling over one hundredth of a mark is a pointless discussion no one can ever win.

At the other end of the spectrum you've got marks out of three or five: the kind of scoring systems that newspapers use to give a general guide, but hopelessly inexact for specialist publications trying to score things right on the money. A reviewer's job, after all, should be to aid someone with their purchasing decision, and, when all your spare cash is riding on someone's opinion, you need one you can trust.

Even then, all review scores are massively subjective and based on a bewildering multitude of individual factors as varied and diverse as the people reading the reviews themselves. Anything from the age of the reviewer to the systems they grew up with to the games played beforehand all have a part to play, and in our long experience even when these factors all collide you still don't all necessarily draw the same conclusions.

Eurogamer will always give a specialist level of expertise when reviewing a game. For example, there's little point giving an RTS to an FPS junkie who thinks shepherding matchstick-men armies around a map is boring, or an RPG to someone obsessed by driving games. And so on. Of course, like most people, all the EGers like multiple genres and very often they cross over - sometimes we have somewhat irrational preferences for one franchise over another - but we're pretty clear with our preferences so it's very unlikely we'll review a game we're not already curious about. Unlike many publications, we try hard to avoid talking about games we can't speak knowledgably about, and if we can't find someone on the team to handle a certain type of game, we'll try and find a freelancer who does know their stuff. The last thing we'd consider doing is get someone who knows nothing about the game's history or the genre to try their luck; the chances are it'll be a pointless exercise.

The actual process of scoring really is not that complicated at all. Clearly if a game is good in the reviewer's estimation it gets a good score. If it's not so good, then we hammer it, and all points in between. So long as we're clear on what those scores actually mean, the only thing left to argue over is the perennial debate over whether person X thinks it's worth more and person Y the opposite.

We'll also aspire to do more than simply identify if something is fun. Knowing why somebody found something entertaining - how the experience made them feel, which elements conspired to do that most effectively, and so on - is far more useful in our estimation than a big number and nothing else. Scores are useful for context, but the review text is equally as important to us, and we hope to you too.

The perfect ten?

A score of ten reflects a game that, within the reviewer's estimation, is something you must buy: this is the message we're trying to convey. On a basic level it's almost certainly the best quality game ever seen within the context of its genre, and that's why Eurogamer doesn't dish them out very often. A score of 10 usually applies to less than a trio of games in any given year.

But all 10s are not born equal. For starters, you might consider that a ten in the RPG genre still isn't as appealing as an FPS that we scored an eight, or be mystified how we could score a football management game a nine when we only gave that survival-horror game you loved an eight. The best rule is to simply rate like with like, and use your own personal taste barometer to gauge whether the genre is of interest to you. Even so, if you're new to a particular genre then something scoring a ten is a very safe bet indeed. As a starting point, the message is you can't get a better game of this type.

Let us make absolutely clear that a 10 is not and probably never will be "the perfect game". There's always something criticisable about a game, however small.

A 10 will inspire the reviewer because it gets so many things correct. It will be something truly groundbreaking and aesthetically successful, be consistently enjoyable, get the balance right in difficulty terms, be technically very impressive, and be polished to a shine. It will leave the player in no doubt that they're playing something special right from the word go and will continue to inspire and amaze throughout. As we've said, this doesn't mean it's perfect, and we'll be sure to say where it goes wrong too, but maybe those niggles are just so minor that you can let it off. Look at anything under a microscope for long enough and you'll see the flaws. But would you kick a supermodel out of bed for farting?

Nine will do just fine

A game scored nine is another must buy for its audience, or at the very least something you should do everything you can to play. People looking for the best of this type of game won't go far wrong with a nine, and it is a practically risk-free purchase. But it is still a score that comes with a few strings attached that we'll do our damndest to wrestle with before we settle on that "so close" of scores.

For example, it may simply be that although it's a flawless game in many respects, it's simply too derivative or not enough of a genre progression to truly warrant giving it the glorious one-oh. Maybe it's too short, or is spoiled by other factors like the voice acting or control issues. Things you can live with but take the shine off it.

We'll have enjoyed the game throughout. Maybe a few sections will have inspired a bit of red mist, but nothing we couldn't handle; by the end of it we'll be sitting with a big fat cigar nodding sagely about our adventures and basking in the afterglow of a job well done.

Maybe it was just familiarity that chipped off that elusive extra mark. There's a certain wow factor you get from playing a truly outstanding ten out of ten game that second, third or even fourth time around that simply isn't there if it's hewn from the same rock, by which point the law of diminishing returns has usually well and truly kicked in.

Many developers and publishers get into a groove within a particular technological generation - particularly since the console market has become so dominant - and games can hit a ceiling of ambition based the limitations of their given engine. It's quite tough to give even the very best games more than a nine if what we're given is maybe only an incremental update.

Eight can be great

Eight is that tricky "almost brilliant" category, that for fans of the genre in most cases will still warrant serious recommendation to go out and buy, while even people into other genres will probably still draw plenty of enjoyment from. You should at the very least try and play the demo, borrow it or rent it first, but bear in mind that it's not necessarily going to appeal to every person who plays it.

There may be stand out moments spoiled by badly judged difficulty spikes, or numerous other minor and niggly factors that gang up and chip off another mark.

Seven not as much heaven

A seven in Eurogamer speak is something we greatly recommend you have a good look at before you buy, because you can bet many highly regarded games over the years have been awarded a seven. If you're a fan of the genre in question then there's a very good chance you'll get a lot out of it, and often you'll hear a game's vocal following doing their best to boost its chances. Conversely, many detractors will seek to kick it down even further. A seven suggests you're into the idea of the game you should still try and check it out - you might like it more than you think, or be able to forgive its less than brilliant visuals. Or maybe it looks good, but the camera issues were just unforgiving, or maybe the AI wasn't up to much.

Over the years, in gaming parlance, seven has become almost the real 'average' score, but this is nonsense that we wish was flushed out of the scoring system with many publications terrified of giving certain games - particularly those of a great significance backed by big ad campaigns - less than this mark. Often giving less than seven is deemed to be the point of no return, with this considered to be the psychological barrier at which point people decide to buy at or not. But we're resistant to the idea and believe seven equates to being a very good score, and one that should be given more credit. In Eurogamer's eyes, it's still a great game, but one that is likely to sport a few areas where the game is blatantly let down - to the detriment of its overall quality and enjoyment.

Six - get your kicks?

A game scored six is 'good', and many of its audience will enjoy it, but approach with caution. It's not good enough to consider rushing out and buying without a fair bit of research first, but worth a rental if your curiosity demands it, and, depending on taste and tolerance of certain issues, it might be something you wouldn't be ashamed to have bought. Six isn't a disastrous score by any means - it's the first score on the way up to represent what we'd deem as a 'good' game, that had the potential to be great but was sufficiently flawed in crucial areas.

More specifically, games which are maybe very derivative (of others or themselves) and nowhere near as good as their rivals but still enjoyable on their own merits will regularly score sixes on Eurogamer. It's the score that shrugs, "Yeah, it's all right, but compared to X it's not technically as polished, the story's so-so, and it's not quite as much fun to play". It could be that it's a quick-fire, cash-in sequel, which is virtually a carbon copy of the previous game with new levels and yet costs just as much. This is something we're definitely not fond of seeing at EG, and although newcomers to the series may argue this doesn't apply to them, we'll try at least make this clear. But we don't review games in isolation; we review in the context of the series and how they measure up against their rivals. Maybe it's the sort of game with a great deal of promise that ended up being released prematurely, and is a little rough around the edges with camera and control issues, technical shortcomings, difficulty spikes that make certain sections harder than they probably should be. You get the picture. Considered by some people as rough diamonds, and will pick up a lot of affection from the more forgiving gamer, but definitely not for everyone.

Five - good enough to survive?

Five is where you really start to fear for a game's quality. It's the score that says "don't buy it unless you're the sort of person who has to have all the games in a particular genre". It's a game that had the potential to be good, but simply ended up saddled with a catalogue of issues that the majority of gamers will not put up with. It's still playable, but the chances are it's so generic and uninspired that you begin to question how it was released in the first place.

A five won't be a disaster. In fact those who won't have played the better games in the genre might even get a great deal of enjoyment out of it. In a lot of cases scores come down to user expectations, but the standards games are built on are constantly shifting sands; what is ground breaking in one generation is the bare minimum standard a few years down the line, and things we once tolerated routinely can suddenly become very irritating. We hear regularly of games built to brutally tight deadlines, and compromises are inevitably made. Sometimes a key member of the team leaves midway through; sometimes it's just down to a lack of talent. But it's not bad as such, it just inspires little but the feeling you could do far better than consider picking up a five out of ten.

One thing to clear up is whether five is the "average". The law of averages suggests that a five should equate to the woolly notion of what an "average" game is worth. We all know this is rarely the case because it'd be very tough to play every single title in a particular genre to really have a true handle on what the average game actually is; and even then it's a subjective reasoning that's inherently flawed. With about 1,000 games being released every year it's a task of unreasonable magnitude to cover even half of them given the size of our team, but we do play a phenomenal number of games between us at EG (over 250 a year), and will attempt to play all the important games in any give field. The chances are, if it's not on Eurogamer there's a good reason why not. Either we're not sent the game, we don't have a person qualified to review it, or we've too many games to review at any one time, and thus have to deal with them in order of priority. But should five be the average? Yes, but it's not an exact science. We'll always try to position a five as our opinion of what an average game represents, but the more significant notion is that it's the score that tells you to approach with extreme caution.

Four - getting into the realms of poor

When you see a four out of ten anywhere you know the reviewer's had a fairly miserable experience and is issuing caution to potential punters to stay away from it. Somewhere there will be some semblance of playability, but it's the kind of game you wouldn't even really want to suggest renting most of the time. At best you should maybe borrow from a friend or have a look in-store (or download the demo if it's a PC game, naturally) if you're really curious and want a second opinion. There are maybe some redeeming features in there; maybe it's technically acceptable in certain areas with some decent graphics, but completely ruined by broken gameplay that's possibly full of bugs, a terrible control system and all manner of collected irritants like bad camera system, offensively bad script and voiceovers, rotten level design, and probably stupid AI quirks or balancing that's too easy or too hard. But games that have scored this badly and suffer from these exact issues often sell very well indeed, so there's no accounting for taste. If you don't like wasting money, then stay the hell away from a four. There are so many better games you could be looking at.

Three - not for me, thanks

Now we're getting into hair-tearing territory, where a game is so flawed in so many ways that you're quite sure that even the developer and the publisher know full well what's wrong with the game. At this point the companies involved often go into damage limitation mode and may not even supply review code to publications in the first place. This is one of the reasons you don't see a lot of games scoring low marks on EG. One of the other main reasons is that there are more than enough good games being released to spend too much time worrying about covering the dross.

A three will be an often quite punishing gaming experience. It will be the kind of game that in all likelihood you can tell is bad just by looking at it. Screenshots may well suggest otherwise, but see it running and you're immediately likely to tell why it's bad without having to get your hands dirty. The controls will almost certainly be shoddy, the playability will be all over the place, crash bugs will probably be evident, graphical glitches, camera problems, nightmarish and arbitrary difficulty hikes that inspire new swear words and quite probably some cringeworthy attempts at cool or the kind of storytelling that ought to be outlawed by an EU directive. You'll definitely want your money back, and you'll tell your gaming friends to stay the hell away, or you'll show them it just to punish or amuse them,

Two - about as entertaining as 'flu

Avoid at all costs - this is less entertaining than setting fire to a ten-pound note. You'll have barely ever seen a 2/10 on EG - and for very good reason. A game this bad almost certainly won't reach our eyes, because publishers generally know better than to send games of this standard to us in the first place. In many cases these are your typical "straight to budget" titles that no sane publisher would try and release at full price, and they certainly wouldn't want us to rip them to shreds in public.

What you're facing here is a game with appalling generic visuals built around an awful design, cursed with cretinous AI, brain-frying audio and controls that feel like they've been designed to upset people or boost sales of replacement game pads. It could just be that the game is just so hideously old fashioned that someone has released the game by mistake. Who knows what goes through the minds of people who feel the need to try and sell crap? Pity them, and pity the fools that stock it and more so the morons that end up buying it without checking first.

One - bring a gun

Only the truly broken games qualify for this, and needless to say you'd have to be on some very good crack to consider parting with cash for anything this bad. But people do. Often all a publisher needs to do is get some two-bit license and it could literally sell faeces in a box. A game that gets a one is probably so fundamentally terrible that the very sight of the disc is offensive to the eyes, and the screenshots hurt your face. Seeing it moving is akin to some obscure torture with awful technical flaws compounded by an almost total lack of playability. It's not just bad, it's actually entered into a new realm of terror, and its flaws will often be quite unintentionally funny. Just run as fast as you can. Very few games this bad ever make it to the shops these days, so you're not likely to encounter them, but they do exist - especially older titles going for suspiciously low prices. Beware.