Duke Nukem Forever exposes crap reviewers - 100 reviews compared

http://www.xentax.com/?p=303

by Dinoguy1000 • 15 years ago

4 4 15 years ago

Steko2• 15 years ago

His methodology?

"We all know that we are right, as the true value of DNF is around 65-70%. Thus, I’d say the 60-90% group of reviewers are The Good."

Srsly?

dirkgadsden• 15 years ago

That's only the beginning of the flaws; I saw no mention of the author taking into account the individual scoring systems that the various reviewers used. Some reviewers can be quite harsh, and would use the majority of the 0-100 spectrum for scoring AAA titles, others (i.e. the ones getting the games for free from the publishers) tend to only use between 65-100 (65 being an arbitrary cutoff) on the scale (they'd give a 65 to a game that another reviewer who used the entire scale would give a 0 to). It's especially ironic that the author bashes Metacritic and other aggregation sites for being "shallow," then proceeds to churn out similarly shallow, infographic/graph-laden content. With something as incredibly subjective and contextual as a video game review, distilling a game to a numbered score is pretty ridiculous; then aggregating those scores together into a totally context-lacking mass is pointless, as the result would be totally lacking in almost any meaning whatsoever.

MrMouse• 15 years ago

The data is what it is. I collected as many scores I could, without the preselection bias that you find with metacritic. Indeed, I do not take into account "individual scoring systems" that various reviewers used, although I did note whether they used a 0-100% scale, or 5 stars. The bottom line is, many a reader will look at the scores to decide, even if you feel that is not the case. Also, reviewers that only use 65-100 are basically incompetent. Use another system if you are not going to use the rest of the scale anyway, or - if you think the actual text of a review matters only - just don't use any scoring system at all.

I present the raw objective scoring data, and these data indicate there are two bell curves possibly, and those bell curves prove the interesting point. I don't care if you project whatever I feel about the game in question on the validity of the data, because that's just nonsense. The data was collected without selection bias, just gathering as much as possible given the timelines. The only thing that matters is that we felt there was something amiss with certain reviews based on our own experience of this game in particular. From then on, it was a tool to expose certain reviewers for showing incompetent behaviour. You can rest assured that I don't care whether DNF sells or not. I do care when reviewers at major sites show questionable conduct. DNF is the tool, from then on we will watch them, also in retrospect. Bad reviews can make or break a game in certain cases, and this kind of power in the wrong hands is not to be tolerated.

Duke Nukem Forever exposes crap reviewers - 100 reviews compared

Related Stories