I’m Zack and this is the Millionth Game Blog on the Internet. Currently this is a solo venture, although the format I’ve come up with could allow friends of mine to weigh in with their own scores on a post, as guest reviewers.
For big publications, a major issue in talking about games critically is the need to come out with a score close to launch. Games take time to play, far longer than it takes to watch a film and type up a review. This means advance copies of games from publishers are often required, leading to an obsequious relationship. And these numbers have consequences; publishers have used aggregated metacritic scores at times to determine a developer’s salary or even to decline funding for a sequel if the game didn’t average an 85/100 score or whatever. That’s probably contributed to the situation where the worst game ever is still going to be scored 6.0/10 if it’s from a big studio.
I think reviews are a good place to go not just to stay relevant and find out about what new games you should be buying, but to learn about good and bad design. In this, there’s value even in looking at a games from 20 years ago that everyone has already played. A comment like “The environments were too small or not varied enough” might make a valid point, and you’ll likely see that kind of comment from me every once in a while. But if the solution in mind is “They should have had a bigger budget,” it probably isn’t anything the developers didn’t already know, and does nothing to educate people to make better games.
As for numerical scores, my belief is that they only provide useful information when understood as one person’s subjective opinion, and only when the reader has a large sample of reviews from that critic on hand. That way, readers can see the big picture of how that critic’s tastes measure up to their own, learning whether or not to trust one reviewer’s opinion with regards to games they’ve never played.
I think a good numerical review scale needs to meet the following criteria:
- It can’t be too arbitrary. Even with a number between 1 and 5, it’s often difficult to decide where to put something. It’s pushing it to call something a 9 as opposed to an 8, and 4 ½ stars is the same thing. With a score like 6.7/10, it’s getting ridiculous. How is a 67 ultimately a better game than a 66, when you have to wade through so many layers of subjectivity to get there? Beyond using a small number as a shorthand for stating that you loved something or that you thought it was just passable, these are useless. I should not think a game was mediocre yet boost it to 4/5 or even 3/5 because of some kind of perceived prestige. And if I hate a genre, say, survival horror, I should tend to give survival horror games a lower score. There’s no pretending I can compensate for my tastes, saying, “I hate this game but I think it’s effective at what it’s trying to do, so I’ll give it a high score.” The score is a measure of my hate, not how well-polished I might perceive a turd as being. And if I really think my dislike isn’t “fair”, I should reconsider writing about it at all.
- A score is personal. If multiple people contribute to the blog, their scores should all be listed separately, never averaged. My idea here is that if I’ve written a full review and then pass the same title on to a guest reviewer, they can update the same post with their own section in a little box at the end, stating their disagreements with the original post, or what they feel was overlooked.
- The whole range of the scale should see use. If you never give anything a 1/5 because you only play mainstream games and believe that a One should be restricted to the kinds of awful shovelware you’d never play or review to begin with, your scale is lopsided. You don’t need as wide a range in saying a game isn’t worth playing, so an “okay” game probably shouldn’t be scored higher than a 2/5.
- Your mileage will vary, but as an additional experiment, I may sometimes mark a post with MMMV: My Milage May Vary. The idea is still to arrive at my personal subjective score, but to note that I may have had the privilege to play with Quality Human Beings that I knew from real life, or to note that I was unable to do so when I had the option. After all, nothing can make a shitty game enjoyable like playing it with people whose company you enjoy. While I wouldn’t go so far as to always cleanly deduct one star from a review to simulate what it would have been like if I had to play solo or with the general public, I think it’s something a reader would want to keep in mind.
I think it would be very cool for a site to have a database of reviews, where you could filter out reviews from a specific person, sort by genre, find all of one person’s 5-star reviews, and so on. I don’t have that kind of technological expertise, but this isn’t that big of a project, so I’ll probably just have a table somewhere of all games reviewed, and a few tags on each post.
Lastly, here’s the rubric I’ve come up with for my own scores. Others may have a few different ideas, but I think this is a good starting place to help assign scores consistently based on what should really matter:
x/5 (unscored): Sometimes a reviewer might not know what to make of a game they played. In this case they could ignore the game, or if it was still interesting and they want to call attention to it, they could write a non-review and give it an x out of 5, meaning unscored. This shouldn’t happen often, and whether it should happen at all is another experimental question, but in cases where the reviewer had an uneasy appreciation for some weird art game, because they were confused about the intent or it was outside of their comfort zone, it’s good to leave the option open for a blank score. The game might otherwise have very boring or tedious elements. It might have made the reviewer uncomfortable. Chances are, if the experience had provoked complicated feelings but was still purely delightful or purely unpleasant, the reviewer would have been comfortable giving the game a 5/5 or a 1/5. But to arbitrarily give a strange and internally polarizing game a 3/5 just out of need for a score, even when it might not come down to a question of design at all? Scores like that could poison the overall usefulness of the system.
Thank you for reading. As Shigeru Miyamoto probably said at one point in time, I imagine: “GAME are made with heart, and we wish everyone will enjoy heart of game with us.”