By using this site, you agree to our Privacy Policy and our Terms of Use. Close

Forums - Gaming - How did metacritic compute some of their reviews????

aikohualda on 31 May 2014
brendude13 said:
aikohualda said:
HigHurtenflurst said:

3.5/5 *100

The problem with that though is the scores don't neccesarily gel together.... Admitedly most (probably like 2/3rds) of reviewers use a stupidly over 'accurate' score out of 100, whether that be literally a percentage or out of 10 but to 1.d.p. makes no difference. As foolish as I think that was of scoring is, something lke Metacritic would be ok if all the reviews used the same scale and metacritic didn't mess with the average in any way. It would just be an average score and while it wouldn't mean a great deal, it could be an indicator of a games quality just due to "wisdom of crowds" effect.

As it is though, many of the reviews included in Metacritic use different scales, and some may use similar scales but have a totally different interpretation of what they mean (though it's rare to find a game reviewer who gives average games a central score in their scale anymore). Metacritic tries to "correct" this using heuristics but it's just trying to find a solution to a problem that shouldn't exist, and they also weight reviews even if they use the same scale anyway.

3.5/5 * 100 = 350/500

it is

3.5/50 * 20 = 70/100

I have no idea what you are doing.

ne he said you are suppose to multiply by 100...

 

it is multiply by 20.... to make it 100....

 

3.5 *20 = 70

5 *20 = 100

 

 

i accidentally put 0 though :P



 

Around the Network
Turkish on 31 May 2014

mediacritic is dumm, dont look at it



HigHurtenflurst on 31 May 2014
aikohualda said:
brendude13 said:
aikohualda said:
HigHurtenflurst said:

3.5/5 *100

The problem with that though is the scores don't neccesarily gel together.... Admitedly most (probably like 2/3rds) of reviewers use a stupidly over 'accurate' score out of 100, whether that be literally a percentage or out of 10 but to 1.d.p. makes no difference. As foolish as I think that was of scoring is, something lke Metacritic would be ok if all the reviews used the same scale and metacritic didn't mess with the average in any way. It would just be an average score and while it wouldn't mean a great deal, it could be an indicator of a games quality just due to "wisdom of crowds" effect.

As it is though, many of the reviews included in Metacritic use different scales, and some may use similar scales but have a totally different interpretation of what they mean (though it's rare to find a game reviewer who gives average games a central score in their scale anymore). Metacritic tries to "correct" this using heuristics but it's just trying to find a solution to a problem that shouldn't exist, and they also weight reviews even if they use the same scale anyway.

3.5/5 * 100 = 350/500

it is

3.5/50 * 20 = 70/100

I have no idea what you are doing.

ne he said you are suppose to multiply by 100...

it is multiply by 20.... to make it 100....

3.5 *20 = 70

5 *20 = 100

i accidentally put 0 though :P

My calculation was correct, it is how you convert a score into a percentage.

3.5/5 = 0.7
0.7*100 = 70

Having said that I probably made it a bit confusing with the space I put after the fraction... but it's still a fraction not a ratio so you don't treat the 3.5 as a separate entity from the 5.

You seem to be working in ratios, which is fine as it works... it just means the calculations change for each different scale if you want to convert it back to a percentage. (for example if it were a score of 3.5/4 you would have to multiply the top & bottom by 25) As Metacritic converts scores into a percentage (which is the same thing as "out of 100") all you have to do is divide the score into the maximum score and multiply by 100.



Stefan.De.Machtige on 31 May 2014

It's just a website. I'm sure they just make that stuff up with the wet finger procedure.



In the wilderness we go alone with our new knowledge and strength.

HigHurtenflurst on 31 May 2014
Stefan.De.Machtige said:
It's just a website. I'm sure they just make that stuff up with the wet finger procedure.

What's the wet finger procedure? (I have a feeling it's a reference to testing for the wind direction but I've not heard it used like that before)



Around the Network
Stefan.De.Machtige on 31 May 2014
HigHurtenflurst said:
Stefan.De.Machtige said:
It's just a website. I'm sure they just make that stuff up with the wet finger procedure.

What's the wet finger procedure? (I have a feeling it's a reference to testing for the wind direction but I've not heard it used like that before)





In the wilderness we go alone with our new knowledge and strength.

Nintentacle on 31 May 2014

 

Biggest brain fart ever.

Guess you can't remember everything they teach you in school!

 

 



HoloDust on 31 May 2014

IMO, translation from stars to points is pretty broken. In my book (and on sites that use it), 3.5 stars is pretty good game, a lot better than what 70 points on Metacritic would indicate.



HigHurtenflurst on 31 May 2014
spurgeonryan said:
HigHurtenflurst said:

3.5/5 *100

The problem with that though is the scores don't neccesarily gel together.... Admitedly most (probably like 2/3rds) of reviewers use a stupidly over 'accurate' score out of 100, whether that be literally a percentage or out of 10 but to 1.d.p. makes no difference. As foolish as I think that was of scoring is, something lke Metacritic would be ok if all the reviews used the same scale and metacritic didn't mess with the average in any way. It would just be an average score and while it wouldn't mean a great deal, it could be an indicator of a games quality just due to "wisdom of crowds" effect.

As it is though, many of the reviews included in Metacritic use different scales, and some may use similar scales but have a totally different interpretation of what they mean (though it's rare to find a game reviewer who gives average games a central score in their scale anymore). Metacritic tries to "correct" this using heuristics but it's just trying to find a solution to a problem that shouldn't exist, and they also weight reviews even if they use the same scale anyway.

Shoot! I thought 3.5 out of 5 was good. Even 3out 5.

It may be. Neither you nor Metacritic know what the 3.5 score given means to the reviewer without reading the review. It may be that the particular reviewer avoids using 5/5 as much as possible due to some ridiculous belief that only "perfection" can get a perfect score (that shouldn't be what scores are for, they are meant for comparing games against each other not against something impossible to create) which means 3.5 is only 2 marks down from the games reviewed as amazing. Or it may be that the particular reviewer truely uses a scale to compare games so that even 2/5 is considered average.
In both those cases 3.5/5 is a good score

People should stop fixating on scores, even aggregated ones.