Analysis of Game Statistical Data using Tableau
Game: Rainbow Six Siege - Operation Velvet Shell (Y2S1)
Tableau Dashboard: http://bit.ly/rainbow6t
Data Sample: bit.ly/r6velvet
DataMining-http://bit.ly/r6_apriori
I have tried to add images in the Tableau to provide better visualization and also added location data from Steam database, as I wanted to plot geographical data but it wasn't available in the original source (hence some data geographical data might not be accurate, but it should serve the purpose of analysis). This can also be referred to a case study for "how to perform data analysis.
Thank you for viewing.
5. RAINBOW SIX SIEGE
Visualization - http://bit.ly/rainbow6t
Data Mining - http://bit.ly/r6_apriori
Data Source - http://bit.ly/r6velvet
6. The foremost important step in
BI/Data Analysis is to understand
the business rules which can also
help to understand data and need
of analysis; from which one can
extract useful insights.
Study
Business
Understand
Data
Define KPIs
BUSINESS RULES
8. Every data is different and must be treated accordingly.
For example, Sales data requires domain knowledge of
Marketing & Finance. Hence one must ensure that data is
interpreted based on domain values.
UNDERSTAND DATA
For this analysis, we would be using Game Statistical data.
Understanding of online gameplay is must here.
10. Before you start with the analysis, one must acknowledge the
need of understanding the database & it’s architecture, because
it complies the integrity & accuracy of the data used for
analysis.
DATAWAREHOUSE OPERATIONS
Simple understanding about the data source and it’s architecture
strengthen the flow of analysis and report generation.
ConnectorsETL
Data
Source
12. Just like any other IT work, there are numerous tools available
to perform the analysis on the data, however it is
recommended to understand the need and capability of
business while finalizing the tools for the operation.
DECIDING TOOLS
Operations
Data Attributes
Price GUI
Usability
Deliverable
13. Following are the tools used for this analysis:
TOOLS
Database
Analysis
Mining
Quick Review
14. Visualizing is the creation of analytics elements like charts, bars and
graph etc. Tableau provides 24 visualizing style natively, one can use
them to create n number of visualization by combining and using them
together.
VISUALIZING THE DATA
You can visualize:
1. KPI
2. Statistical data
3. Any information which is given as measure & dimension in the
data
15. What are KPIs?
Key Performance Indicators are the critical indicators of progress
toward an intended result. KPIs provides a focus for strategic and
operational improvement, create an analytical basis for decision making
and help focus attention on what matters most.
“What gets measured gets done”
KPI
16. How to define a KPI?
▪ Is it measurable?
▪ Is it an important aspect of business?
▪ Does it influence the outcome?
▪ Capable of forecast?
▪ Accurately defined in the data?
DEFINE KPIs
Example: Skillrank
19. The insights are useful information which can suggested the state of the
data which could not be covered using analysis. Data mining is the
technique used to get insights.
1. Association Mining
2. Tracking patterns
3. Classification
4. Outlier detection
5. Clustering
6. Regression
7. Prediction
GETTING INSIGHTS
20. To analyze the pattern of loadout
selection (Operator, Primary
Weapon & Secondary Weapon)
association mining was carried out
on the complete dataset. Since
market basket analysis could not
be applied on such data but a
close and similar analysis was
done using Apriori Algorithm.
ASSOCIATION MINING
22. ASSOCIATION MINING
I used Python along with SciKit Learn to carry out the mining operations.
Use algorithm in your program, and provides values from data.
Let it run and complete…
23. ASSOCIATION MINING
Once you got the result, visualize this result to understand mined
information:
CSV
Tableau
25. Clusters show the Win ratio between countries based on their
average playtime and number of Kills achieved in a match.
CLUSTERING
26. Time series analysis comprises methods for analyzing time series data. This can be
extended to all three Platform and its comparison for distinguishing the performance
and online activity of Players on the same.
TIME SERIES ANALYSIS
27. Reports in DA/Tableau means the representation of the
visualized data presented in form of Dashboards.
GENERATING REPORTS
28. This dashboard provides the demographics stats of the data available; distribution overview for Platform, Region
and Game Mode
PLAYER DEMOGRAPHICS
29. Weapons are key point to win a match. This dashboard represents the stats of the Weapon data applied
using a Filter.
WEAPONS DASHBOARD
30. RECOMMENDATIONS
What are Recommendations and what do they mean in Business
Intelligence?
Providing suggestions or decisions by carefully looking at the analysed
data is a Recommendation.
Recommendation summarizes the work done by analysis, and is the next
step towards improving the business.
If not said, always provide not less, not more but 3
Recommendations.
31. Following are the recommendations made on the analysed data:
1. PC platform lacks the user base by ▼25%. To increase the players
on PC platform, XXXXX actions must be taken.
2. Game Modes are also observed under unequal balancing. PvP-
Hostage to be specific which currently have only ▼12% of acquired
player taste.
RECOMMENDATIONS
32. RECOMMENDATIONS
3. Comprehension of selecting exact Operator (character in the game)
behavior is impossible, since everyone is an individual, it is useful to
get the Operator with most & least number of Win & Kill Ratio.
It was performed using association mining; which concludes the
most picked Weapons by each of the Operator based on higher
Win/Kill ratio.
Jager, IQ & Bandit Operators are the least performing ones.