One of my final homework assignments for CS 7250, Information Visualization, was to create a visualization on something personal. Gaming is one of my major hobbies, so I gathered data on every video game I’ve played for at least one hour according to Steam’s built-in metrics (this data was gathered on November 18, 2025). Aside from just hours played and date last played, there were also achievements to consider, the genre of the game, whether or not a game is “randomized” (different every time you play, like a board game), and some other miscellaneous data I ended up not using (such as “could I run it on my current computer with little effort” and “how many times have I bought this game”).
The most challenging part was figuring out how I wanted to do a rating system. There were a lot of games I hadn’t thought about in almost 10 years, and I didn’t remember anything about them. Instead of doing a 1-5 “how much did you like this game” scale, I decided instead to do a yes/no/maybe “would I play this game again?”, and considered that for each game.
The Vis
Color selection tool is below the vis. Select any bar in the bar charts or color icon in the legend to show only that type of game, or brush over the points in the scatter plot to limit to a certain amount of hours played or achievements completed.
Some fun insights I gained about myself and my gaming habits from this project:
- Despite being the largest genre within my played library and the highest average achievement completion rate, puzzle games have the second least amount of time spent playing and the lowest average hours per game of all the genres. I do think this is easily explained by the fact that once a puzzle game is solved, there’s very little replay value for me unless there’s some other draw like an interesting story or a particularly impressive visual style. The games that are classified as puzzle and “would replay” are all story-driven puzzle games (point-and-click adventures, essentially) that do something interesting, like Return of the Obra Dinn with its stunning art style and framing device.
- I have put a lot more time into games with randomized runs, particularly strategy games. About a third of my overall gameplay time is in randomized strategy games, which I’m unsurprised by given my love of card battler/deck-building games like Slay the Spire. The trend of a game with randomization having more hours than a game of the same genre without randomization holds true for all but simulation games and puzzle games. Puzzle games it’s unsurprising, since there are very few randomized puzzle games worth playing; the two I have are a word game and a multiplayer puzzle game. Similarly, there are not many simulation games with randomization. The genre with the starkest difference is action, where randomized games have a mean play time of 68 hours while non-randomized games have a mean play time of 22 hours.
- I have put more time and had higher completion rates for games I thought I would replay, but this does depend on the genre. Only the action and strategy genres have more play time in the “replayable” games than in the “non-replayable” games, but if you combine the two, the replayable games in those genres represent more than a third of my overall play time at over 1200 hours. This is enough to make the “yes replay” category have a higher total play time than the “maybe” and “no” categories.
- I tend to play a lot more new games for a short amount of time in the beginning of a year, and fewer towards the end of a year. January and March are the two most common months for me to stop playing games, which have typically correlated with winter and spring breaks for when I was in school. Many of these games had a low play time, meaning I would put a few hours into a new game before moving on.
Design Choices
With the scatter plot, I used a log scale for hours played to spread out the points across the whole chart, rather than having most bunched up along the left side. Additionally, the size of each point correlates to the number of earned achievements. When I was gathering the data, I noticed there were a lot of games where I had 100% completed them, but they had only 10 achievements. Comparatively, there were some games in the 40-60% completion rate where I had over 200 achievements earned.
For the timeline charts, I did initially want a toggle between the three time units, but I was unable to figure that out (still working on that problem!). I used an area chart rather than a line chart because I wanted the visible bands of color, rather than heavily overlapping lines. Visually it’s much cleaner, and also allows me to see the overall trends. However, the drawback of this is that if I select very few points, it doesn’t show anything interesting and just becomes a full bar of color.
Color-wise, I decided to use three different encodings to highlight different data I wanted to know about: genre, randomization, and replayability.
- Genre feels like an obvious choice to examine, since that’s often one of the first things I check when deciding if I’m going to buy a new game or not. (I’m pickier when it comes to RPGs versus strategy games, for example.)
- When a game is randomized (so that every round or whatever is different than the last), I feel like I will put more hours into it because it stays fresh longer, so I wanted to examine whether this feeling was true or not (it is).
- I find that there’s some games I go back to a lot, and examining if there were any trends in that was interesting.