Let’s recap VALORANT Champions 2023 in data. [With prompt examples]

Published in

Morph Blog

6 min readSep 8, 2023

Hello there! In this article, I will show you an example of a no-code data analysis using Morph.

This time, I will focus on stats from a popular E-sports world championship that took place recently (we have some E-sports watchers in the Morph team, so it was fun to do this 😂).

What is VALORANT Champions 2023?

VALORANT Champions 2023 is the world championship of the globally popular FPS game VALORANT and is the culmination of the 2023 season.

Sixteen teams from around the world gathered to compete for the title after winning regional qualifiers.

Tournament highlight:

VALORANT Champions 2023 through the eyes of data

In this article, we will look back at VALORANT Champions 2023 from the perspective of data, based on each player’s stats throughout the tournament.

Interesting look at individual stats

When analyzing sport of any kind from data, there are many different perspectives one may take.

VALORANT is a 5vs5 tactical shooter with 8 maps and 19 characters. Therefore, the tactical aspect is very important, and it is interesting to analyze the maps and stay abreast of recent developments in the game.

I will cover individual stats in this article for the following reasons:

It is difficult to talk about tactics and game development from data. As is the case with data analysis in regular sports, it is very difficult to convert the ever-changing game situation in a three-dimensional space into data. It is also difficult to talk about complex match situations from a data perspective, since factors that are difficult to capture, such as the psychological state of the players, are also likely to play a large role. Individual stats would be easier to work with.
Extraordinary plays. In soccer, baseball, and other sports, there are times when a super play occurs that makes you wonder how such a thing is possible. I think that these kinds of phenomenal performances by individual players are easy to capture as stats.

In e-sports, especially in shooters, player skill can be quantified in things like reflexes and mouse control accuracy. Since we ourselves use a mouse and keyboard every day, it’s eye-opening to see how far they can push these everyday devices, and the dexterity on display.

Indicator: KD

KD (also written K/D or KDR) is the Kill/Death ratio, which is the number of opponents killed divided by the number of times the played died.

The value depends on the rules and trends of each title, but the higher the value, the more often the player wins.

Analysis

Now, without further ado, let's look at the actual stats.

For a step-by-step guide to running a data analysis using Morph's Notebook, please refer to this article.

The actual prompts used are also included at the bottom of each analysis.

Top 10 Players by K/D

Prompt:

Visualize the top 10 players as a horizontal bar graph, sorted by increasing value of kill_per_map / death_per_map.
The label for each series should be player_name [team_name].

The color of the chart should be dark blue.

Also, indicate the average kill_per_map / death_per_map for all players with a red line in the chart.

Here are the top 10 KD players. The top place goes to Demon1 from Evil Geniuses, but hot on his heels are Alfajer and Leo from Fnatic, even though their team came in 4th place in the tournament.

Team KD Average

Prompt:

Visualize the top 10 players as a horizontal bar graph, sorted by increasing value of kill_per_map / death_per_map.
The label for each series should be player_name [team_name].

The color of the chart should be dark blue.

Also, indicate the average kill_per_map / death_per_map for all players with a red line in the chart.

Here is a visualization of the average KD per team. As expected, the top teams have higher scores!

Plot of median and variance of KD

Prompt:

For each team_name, show the "variance of the player's kill_per_map / death_per_map values" and the "average of the player's kill_per_map / death_per_map values" in a scatter plot.Each dot should be labeled with the team_name, which should be in text only, with no background color.The vertical should be the mean and the horizontal the variance.You do not need to show precedents.

This is an analysis with a different perspective. The horizontal axis is the variance of KD between the 5 players on the team, and the vertical axis is the median KD of the team.

In other words, the higher you go, the higher the KD, and the further to the right, the greater the dispersion among players.

The top teams naturally have higher KDs, but while Fnatic in 4th place, for example, has a large variance, Paper Rex in 2nd place shows a very cohesive performance.

Team KD + Best and worst KD player in a team

Prompt:

Show the average of the kill_per_map / death_per_map values for each player for each team_name as a horizontal bar graph. This bar graph should be dark blue. 
Also, show each player's kill_per_map / death_per_map value as a dot, overlapping the bar graph. The dots for the highest and lowest kill_per_map / death_per_map player on each team should also show player_name. 
No precedence is required. These dots and player_name should be bright orange.Also, the average kill_per_map / death_per_map of all players should be shown as a red line in the chart.The entire chart should be sorted in descending order of the average kill_per_map / death_per_map value of the players.
The vertical line should be the team, and the horizontal line should be the average of the kill_per_map / death_per_map values.

Here is a plot of each player’s KD as dots on a bar graph of each team’s average KD.

You can see at a glance that Demon1 on team Evil Geniuses is prominent on the team. Also, Paper Rex is still cohesive.

This kind of complex visualization is very difficult to build with charting tools or BI tools, but with prompts, it can be achieved with detailed instructions!