Is a 5-Point Scoring System Enough For Games?

[Ivo]

5 stars should be more than enough, IF you have the review. In fact I believe there is no real need for ratings: a review is an opinion and if you want the opinion you should read it. Unfortunately many like to just look at the score, and in that aspect, without a text review, I believe even a scale with 200 discrete points is insufficient (it helps when there are more scales, like separate scores for different aspects).

Anyway as I understand the idea of the database, you can make it from 0 to 20 with decimals if you want, or %, if you rather, if you think 5 stars is too little. The numbers you are going to put in any scale are going to be more or less made up anyhow.

[Gamerforlife]

I was leaning towards a ten point scale when I read this, but they're kind of pointless if you think about it. Anything lower than a five is crap, so who cares if it is 1,2,3 or 4? Putting decimals in there is even more pointless if it's under a five. So if you take out numbers 1-4, that leaves six numbers. Might as well just go with a simple five star system. Don't know if that makes sense to anyone. I'd maybe use something like this personally:

One star=garbage
Two stars=average, just a game that doesn't suck, but doesn't make an impression either. Rental fare for genre fans starved for a new game.
Three stars=Good game. Fun to play, and maybe even has some originality and good ideas. Has some aggravating flaws though. The kind of game that maybe could have used a bit more development time for some polish and refinement, or maybe a slightly more talented development team. These are the kind of games that are going to make you smile just as often as they piss you off.
Four stars=Great game. Very high quality. Whatever flaws it has are pretty forgivable given how outweighed they are by the pros
Five stars=Fantastic. Cream of the crop. Nearly flawless. In fact, whatever flaws it does have can arguably be viewed as personal gripes that are subjective to each individual player. This does not represent perfection, because in my opinion there is no such thing as a perfect game, even though I have played some that come pretty close to this mythical status.

[executioner]

do a poll on this one... 5 or 10

[The Apprentice]

I prefer a percentage from 1-100. That or leter grades.

[Curlypaul]

Everyone seems to have a totally different opinion on this so heres mine:

I never saw the point of a hundred point scale because like already mentioned arguing that one game is 1% better than another is just silly. I only ever look at the first number of the percentage anyway and like Mozgus suggests on a 5 point scale, a game scoring 5 is equal to +80% and the game is worth your time.

I remember years ago one of the major Ninetendo magazines gave Mario64 100% and just remembr thinking how utterly stupid that was. Apart from the fact that next year a game even better could be released (then what are they supposes to give that game? 110%) there is no such thing as the perect game, it had an irritatingly stupid camera, and perhaps more importantly - I dont like platformers - a fact that technical wonderment at the lush polygon world and the analogue stick (I had already played nights to death) is not likely to quell. Giving a game 10/10 suggests that the game is generally excellent at what it does, but leaves space for some minor flaws to be explained in the text.

Which leads me nicely to the scale of my choosing - a 10 point scale. I also prefer Gamestm scale of scoring. Rather that a -7 being 'ok if you are a hardcore fan of the genre' and everything below it being crap and everything above it being ok they take the more sensible aproach and give a 5 to the games that are average. That way the scale allows for equal grades of crapness and greatness.

[ott0bot]

Five stars are plenty, 1-10 scale is good too. If you want to solely judge a game by the number of stars instead of reading the review and assessing whether or not it the stlye of game you like, you are probably live to close to a power plant or your parents may be related.

If this is a user based review system, remember, we have the voting power. Also, if they are site based reviews you need you have fair and neutral reviewers (which if you've been reading this site for a while you shouldn't expect anything less from Racket). Some of my favorite games games got terrible reviews because the graphics weren't "next gen" enough because at the time that was what was desirable. Most of the games that will be reviewed have had time to be recognized as classics or their replay values can be fairly assessed.

Just leave it to Racket, he's not going to allow some crappy biased GamePro review system to be put in effect. Although those guys with the thumbs up and spikey hair were kinda sweet.

[racketboy]

Eventually.... (not sure when), I would like to implement a feature that lets users give a thumbs-up or thumbs-down to individual reviews. reviews that have more "ups" will be weighted more and ones with "downs" will obviously be weighted less.

[Niode]

Like a digg style system? That sounds promising.


I prefer C&VG 1,2,3,4, thumbs up scale. I've always enjoyed their reviews and it turned me onto a lot more games that got 4 or 3 out of 5, whereas if they had got 70% or 80% i would've probably gave them a miss. It's such a shame they folded the magazine

I vote for 5 point scale. I relate more to the scale. % is stupid. What seperates a game that got 89% from one that got 90%? is the 90% 1% better or something? I just don't get the need for the details. You might as well just round up and give it a 9. If you're going to do that you might as well just have a 10 point scale.

[racketboy]

The interface would be like the Digg comment system, which other sites have adapted as well.

And I like your philosophy as well -- it's well-stated.

[Niode]

That being said I wouldn't be too disheartened if it was an out of 10 scale, but I think that out of 5 seems to be more fair. If it was out of 10 you just get people voting poor games at 7/10 6/10, I just read those as 1 or 2 out of five as very few games these days are truly abysmal, most games have at least some redeeming features.

I guess i'm just a little miffed at modern magazines/sites rating out of 10 or % and giving poor games 5(0%) and good games 9(0%). It seems silly to have such a wide range scale and only using 50% why not simplify and out of 5?

I think that when you have out of 5 rating you have more scope to give games ratings of 1 or 2 if they are poor (as you have less scope and flexibility for rating). However that being said how do you rate a game 0. By starting at 1 every game must have some merit as they are 1/5 as good as a 'perfect' game (mathematically speaking). I know it doesn't always work like that but I can see merits on both sides of the argument.

Merits of 5 based scoring

- 4/5 out of 5 are both classed as good quality games and worth your time.

- 5 has less stigma about it being a 'perfect' game (well I know i don't deem 5 out of 5 as perfect. ie 100% gives the impression that it's perfect, as in 100% = nothing wrong)

- less numbers encourages ratings across the scale rather than stick with the modern trend of games being rated 7/10 as being poor.

Bad points about 5 based scoring

- due to there being less numbers, ratings do have large gaps so it's hard to distinguish the merits of games rated 3 or 4 out of 5. (double edged sword that methinks)

- hard to give games ratings of 1 or 2 out of 5 as they both seem to be as poor as each other.


I guess theres really good and bad points to both. I would strongly suggest against % based rating systems though. I guess the only way to come to a conclusion is to have a poll on the choice of 5 or 10.