<- read_csv("vgsales.csv")
dataVG <- dataVG %>%
dataPlatformFreq select(Name, Platform, Genre, Global_Sales)
<- transform(dataPlatformFreq,"Platform_Frequency" =
dataPlatformFreq ave(seq(nrow(dataVG)), Platform, FUN = length)) %>%
group_by(Platform)
Exploratory: ‘Best’ Video Game
When it comes to hobbies, few would rank above video games. The synergistic mix of audio and visual design give it the most amount of versatility. From casual single screen puzzle games to open world, story rich experiences that give the player freedom to decide how they’d like to enjoy it. The data I’m working with is titled Video Game Sales from Gregory Smith in kaggle, it compiles sales data from more than 16,500 games.
With it I will attempt to answer a couple of questions:
What is the best video game? Dependent on console? Dependent on genre?
There are many different ways to classify a game as ‘good’ many of which are not included in the data, this isn’t going to be a commentary of the quality of these games, just their commercial success. Several of these titles were sold multiple consoles, their sales would the total between them.
After cleaning the data a bit we could compile the unique values under “Platform” to show how many games are in the data with respect to their platform. We could also calculate the total (millions) of sales those games had. Realistically since I’m simplifying my goal to find the ‘best’ I’ shouldn’t limit myself to the various regions the data provides, such that I’m not taking into consideration the most sold in the individual regions the data has (North America, European Union, Japan), Global Sales are the most important to me in this exploratory analysis.
Platform ...Total_Games ...Total_Global_Sales
1 DS 2163 822.49
2 PS2 2161 1255.64
3 PS3 1329 957.84
4 Wii 1325 926.71
5 X360 1265 979.96
6 PSP 1213 296.28
7 PS 1196 730.66
8 PC 960 258.82
9 XB 824 258.26
10 GBA 822 318.50
11 GC 556 199.36
12 3DS 509 247.46
13 PSV 413 61.93
14 PS4 336 278.10
15 N64 319 218.88
16 SNES 239 200.05
17 XOne 213 141.06
18 SAT 173 33.59
19 WiiU 143 81.86
20 2600 133 97.08
21 GB 98 255.45
22 NES 98 251.07
23 DC 52 15.97
24 GEN 27 28.36
25 NG 12 1.44
26 SCD 6 1.87
27 WS 6 1.42
28 3DO 3 0.10
29 TG16 2 0.16
30 GG 1 0.04
31 PCFX 1 0.03
Above, I thought it necessary for clarity to see the data in descending order with respect to the total amount of titles per console in the data set. As you would notice, there is a large difference between the quantity of titles the consoles have. To mediate this graphically I separated the data into three sections, above 250, below 250 but above 25, and below 25 games.
I decided to include the Genre ‘amount’ in the graphs above to get a general idea of which categories are more prominent for each console. Genre’s would have a dedicated graph later on.
Lets now get what we would call the “best” game with respect to consoles:
Many of these titles are exactly what you’d expect, Super Mario Bros for the Nintendo Entertainment System (NES) and Pacma for the Atari 2600. Although there are some notable titles I wasn’t expecting to find like Minecraft selling 2.25 million copies for the Playstation Vita (PSV).
Name ...Platform ...Global Sales
1 Pac-Man 2600 7.81
2 Policenauts 3DO 0.06
3 Pokemon X/Pokemon Y 3DS 14.35
4 Sonic Adventure DC 2.42
5 New Super Mario Bros. DS 30.01
6 Pokemon Red/Pokemon Blue GB 31.37
7 Pokemon Ruby/Pokemon Sapphire GBA 15.85
8 Super Smash Bros. Melee GC 7.07
9 Sonic the Hedgehog 2 GEN 6.03
10 Sonic the Hedgehog 2 (8-bit) GG 0.04
11 Super Mario 64 N64 11.89
12 Super Mario Bros. NES 40.24
13 Samurai Shodown II NG 0.25
14 The Sims 3 PC 8.11
15 Blue Breaker: Ken Yorimo Hohoemi o PCFX 0.03
16 Gran Turismo PS 10.95
17 Grand Theft Auto: San Andreas PS2 20.81
18 Grand Theft Auto V PS3 21.40
19 Call of Duty: Black Ops 3 PS4 14.24
20 Grand Theft Auto: Liberty City Stories PSP 7.72
21 Minecraft PSV 2.25
22 Virtua Fighter 2 SAT 1.93
23 Sonic CD SCD 1.50
24 Super Mario World SNES 20.61
25 Doukyuusei TG16 0.14
26 Final Fantasy WS 0.51
27 Wii Sports Wii 82.74
28 Mario Kart 8 WiiU 6.96
29 Kinect Adventures! X360 21.82
30 Halo 2 XB 8.49
31 Call of Duty: Black Ops 3 XOne 7.30
The following graph shows how all the titles are distributed based on their genre.
Lets now get what we would call the “best” game with respect to Genre:
Similar to the previous table an expected title to see is Grand Theft Auto V under Action. I would have considered the shooter category to go to one of the Call of Duty titles but I suppose that was my recency bias. Duck Hunt ends up being the most sold game for the “shooter” genre with 28.31 million copies sold globally.
Name ...Genre ...Global Sales
1 Grand Theft Auto V Action 21.40
2 Super Mario Land 2: 6 Golden Coins Adventure 11.18
3 Super Smash Bros. Brawl Fighting 13.04
4 Wii Play Misc 29.02
5 Super Mario Bros. Platform 40.24
6 Tetris Puzzle 30.26
7 Mario Kart Wii Racing 35.82
8 Pokemon Red/Pokemon Blue Role-Playing 31.37
9 Duck Hunt Shooter 28.31
10 Nintendogs Simulation 24.76
11 Wii Sports Sports 82.74
12 Pokemon Stadium Strategy 5.45