Amazon reviews from the PS4 and Xbox One next-gen console wars

Tweet about this on TwitterShare on FacebookShare on RedditShare on Google+Email this to someonePin on Pinterest

The Xbox one has now been out for a couple days – which has given it enough time to gain a fair number of reviews on Amazon. The same goes for the PS4 which has had quite a bit more time.

However taking a look at the them reveals that a significant number of scathing 1* reviews are not considered “verified”, which means that the person didn’t actually buy the product from Amazon. Now, that doesn’t mean those are all fake, but it’s pretty easy to see that die-hard fanboys from both sides of the fence are attempting to create bad rep for the other.

I was interested to work out how many of these reviews were just created by trolls, so I wrote a bit of code to scrape all the reviews (only from and then attempted to make sense of them in the graphs below.


You’re welcome to work out other stuff from the data too if you’d like. I’ve formatted the data as JSON and uploaded the PS4 data here and the XB1 data here.

Tweet about this on TwitterShare on FacebookShare on RedditShare on Google+Email this to someonePin on Pinterest
  • __m

    so in other words: xbox one sucks

    • RogueCode

      Huh? How do you figure that?

    • jj

      yep sub hd gaming for $500

      no thanks

    • Youareatroll

      The irony of this comment

  • Ted

    I often wonder what causes this strange phenomenon where, I like A therefore I must trash B and all associated with it. Mac v. PC, Nikon v. Canon, any two cars, the list goes on. Very strange.

    • jf

      Same. I believe it has to do with justifying that you made the right decision by bashing the other in public. At least initially. Then hyper brand loyalty kicks in and it becomes a religious war.

  • derp

    Why use pie charts? Your data is one-dimensional, not two-dimensional, so it’d be better to use bars. Plus, since bars are more compact, you can display the data side by side, allowing for an easier comparison.

    • RogueCode

      Personally I find the data as easy to compare in these charts as it would have been in a bar chart.

      I also provided the raw data so that someone who is much better at analytics could produce something.

      • Jauny

        I agree, those pie charts look nice!



  • This page is unreadable in mobile browsers.

    • RogueCode

      Sorry about that, what phone do you have?
      Works fine on the ones I have tested.

  • kftgr

    Wow, that 1* stat comparison is so telling and nicely sums up the whole trolling-PS4-fanboys trend that has been plaguing the gaming sites for most of the year.

    BTW, to be fair, can you update the numbers on Dec 1 and compare to the Nov 1 PS4 numbers? This would be a comparison based on the same number of days available. More people would have received their Xbox Ones and have more reports of defects. I’d gather that would make the results less one-sided, but the stark 1* review difference would still be prevalent.

  • coip

    I don’t understand the charts. For instance, the first one looks at source of 1-star reviews. The legend on top has black as XB1 and white as PS4. Then below that, the legend says that teal is for verified reviews, black is for unverified, and white is for reviews from trolls. So, is white for PS4 or for trolls? Is black for XB1 or for unverified reviews? It doesn’t make sense. Further complicating things, in the center of the first graph it says XB1. Is it only for XB1? Then why is PS4 listed in the legend? What does the graph mean: that Xbox One unfairly received more troll reviews?

    Same goes with the second graph. It says black is PS4 and white is XB1–the opposite of the first graph, by the way, which needlessly confuses the reader–but the graph only shows black and teal. Then below the graph we see that black means unverified and teal means verified reviews. And then in the center it says PS4. If it’s only for PS4, why does the legend above it list XB1? It’s not clear what the graphs are measuring.