Becoming an Esports Analyst – Analysts in Esports Explained

 

 

In esports, esports team analysts are a massive asset to the esports teams that carry one in a tier 2 and tier 1 setting. But you may wonder what is an esports analyst.

An esports analyst works closely with the team’s coach to determine specific mistakes made by players by closely observing game replays and analyzing the statistics of the team’s performance relative to their competition. The esports analyst will watch certain games frame-by-frame and generate his own statistics of players’ performance when needed.

A good esports analyst can make all the difference in esports. In this article, I will be breaking down exactly what an esports analyst does, their responsibilities, if becoming an esports analyst is right for you and the process you’ll need to go through to become a paid esports analyst.

Note: we are talking specifically about analyst that works with an esports team in this article, not ones that work with the publishers or major tournament organizers,

What is an Esports Analyst

An esports analyst isn’t a necessary position in an esports team for most games, yet it’s a valuable asset that could make the difference between a good team and a great team.

This individual would work directly with the esports coach to ensure the team is as successful as possible. What makes the coach different from the analyst is that the coach helps the team with soft skills, mindset, mentality, communication and personal issues. The coach also helps develop strategies and resolve very obvious mistakes by the player, but the analyst takes it a step further and micro-analyzes the game.

The analyst will spend a majority of his time watching replays of games, typically frame-by-frame in certain games like FGC, and running the data generated by the game’s internal systems. Sometimes the analyst may even need to generate his own analytics or use third-party software to generate it, similar to physical sports.

The analyst must communicate these findings with the coach and the players to help rectify these issues. These may include something as specific as to their KDA at a certain period of time is a few points lower than the industry standard.

Person Holding Blue and Clear Ballpoint Pen

(p.s. we have actual good pictures below, this is just to appease the Google SEO gods 🙂 )

Or it may be pointing out a trend, such as when the team is ahead by 15 kills or 12 health points, on average the team tends to lose 64% of the time. A weakness of overconfidence may be in question at this point, or it could be the strategy employed in getting ahead early fails to close out a lead (or in this specific example, could result in closing out a perfectly winnable game harder than it should be).

This provides insights on areas of improvement to the team that aren’t obvious or even noticeable without the congregation of data from a multitude of games from the team and other teams at a microscopic level.

Now that we talked about the value of the analyst in esports, let’s talk a bit about the exact workload of an analyst.

Responsibilities of an Esports Analyst

An esports analyst’s responsibility is straightforward but not simple. Their primary objective is to look at trends and underlining imperfections that are holding back the team’s ability to perform at its highest level.

That said, let’s run through the jobs accomplished by the analyst. Below is a summary list of their jobs (with a further breakdown below it):

  • Watch through the gameplay to determine mistakes. In reaction speed games, such as FGC games, a frame-by-frame analysis may be required.
  • Go through the game’s internal statistics and analyze trends. This would include analyzing the team’s scrim games, tournament games and a comparison with other team’s statistics to compare industry standards.
  • Generate their own statistics where the need is or using a third-party software to find additional statistics. This data should only be information that provides overall value to not incur waste.
  • Communicate trends to the coach and team. This would be useful insights on pitfalls and downfalls for the team in varying situations. The usage of strong visuals, such as picture diagrams, infographics and graphs is required.
  • In higher-ranked teams, also provide insights on the weaknesses of the opposing team that the analyst’s team may face off against.

That is an overview of the responsibilities. To give you a more job posting style description of the responsibilities, below is a list of responsibilities for a Head Analyst for League of Legends at Team Liquid:

Let’s break down the functions with a basic example. If you want to skip over to the next section where we breakdown who is right to be an esports analyst, click here.

Practical Example of the Life as an Esports Analyst

For this example, I’m going to give a very basic overview in League of Legends. If we wanted to analyze an individual game, an esports analyst would want to run through the game’s video replay built into the game client. They would want to pay extra close attention to team fights and lane skirmishes to determine critical errors by the team.

Moreover, the analyst is particularly wanting to look at the backend data. In LoL, the data is presented by the client itself. Below is an image of how that information looks. The analyst may also generate his own data if the necessity is there.

For example, recording the number of times the support roams and see the correlation to winning the bot lane and games overall. Since the analyst is getting paid as an employee (so based on hours), they have to ensure any data they gather manually has business value to justify the time invested.

Next, the analyst would pull that data into their own database. This may be an SQL database connected to their backend servers, a database associated with a specific application they use for analytics or something as simple as a spreadsheet as shown below.

Note that the picture below shows all the stats for League of Legends players in the regional tier 1 tournament leading to the 2020 worlds championships. I chose to show this picture simply because I had this on deck (for my own analytic work I did previously), but the analysts would likely work with data only on their team THEN contrast that to other teams.

With that information, the analyst can determine correlations between various objectives to determine a few items. These items include what factors contribute to the team’s highest success, which factors are the weakest or least consistent of the team or how their team may vary from industry standards (known as the standard deviation in the data analytics world, even outside esports and sports!).

Ideally, the analysts want to make graphs to communicate the strengths and weaknesses of the teams. The visual explanation far outweighs this data dump you see above, especially when communicating to the team administration (for business value) and the players (for performance enhancements). An example would be the image below (note, this is an outdated image from League of Analytics in 2016):

The presentation could consist of a topic such as “Which top laner should we recruit” where the analyst was scouting and generating data to determine who would best fit the void. This presentation would be shown to business administration. Another presentation could be about “How much to pay our new ADC, Power of Evil”, also shown to business administration. An example for a presentation for the players and coach would be “Team Fight critical errors” consisting of in-game screenshots and graphs showing how certain players engaged too early or didn’t communicate properly certain cooldowns/positionings.

That is a rundown of what an esports analyst is actually doing. Obviously, this process requires a lot of time investment and creativity on the analysts’ part. How does one know what data to look at versus another? You can use mathematical approaches such as the correlation between a data set to the overall win-rate. This is ideally done with all the teams’ data combined and it will show you what is called a correlation coefficient (analytics term also used outside of sports and esports).

Now, for the moment you were waiting for!

Who Should Become An Esports Analyst

Now that we got the boring stuff out of the way, let’s talk about if being an esports analyst is right for you! There are a few factors that play a role in this decision, below is the high level of just a few:

  • Enjoy and are skilled with numbers.
  • Creative and have the ability to analyze potential trends by looking at raw numbers (to later be run through a computer program to confirm correlations)
  • Enjoy the game they are an analyst for. Including watching and rewatching the same replay a multitude of times. Even to the point of watching frame-by-frame and generating your own data when needed.
  • Good understanding of the competitive scene of their game. This can be learned, but it’s better to be an enthusiast of the esports scene as you’d be expected to watch a lot of games during work and in your own time to best see your team and the competition in action.

I’ll break this down further below. Before I do, for reference on how top esports team describes a strong candidate for their esports analyst position, check out this image from the same Team Liquid job posting below:

They basically say the same things with some regular business stuff such as “taking risks” and “able to communicate”. Now, to breakdown further the qualifications. If you are already solid on being an esports analyst and you want to learn how to go about doing so, click here to skip below

Your Ability with Numbers

How good are you with numbers? As an analyst you watch videos, but you aren’t really getting paid to do that. That’s primarily what the coach is there to help players with, your objective is to watch exactly what will contribute to the numbers, but your main function is to play with said numbers.

Let me reframe that question, how do you feel when you see this data dump below:

Player Position Games Win rate KDA Avg kills Avg deaths Avg assists CSM GPM KP% DMG% DPM VSPM Avg WPM Avg WCPM Avg VWPM GD@15 CSD@15 XPD@15 FB % FB Victim Penta Kills Solo Kills
Bruce  ADC 6 17% 1.7 1.3 3 3.8 9 389 59.10% 29.1 477 1.07 0.4 0.38 0.11 -188 3 -164 0% 0% 0 1
Deft  ADC 6 67% 7.9 2.2 1.2 7 9.6 438 59.80% 25.5 511 1.15 0.51 0.26 0.18 766 17 530 0% 17% 0
Doublelift  ADC 6 0% 1.3 1.8 3.5 2.7 8.4 359 50.60% 28.3 445 1.09 0.38 0.33 0.16 -616 -9 -342 17% 17% 0
Gadget  ADC 6 0% 1.4 2 3.5 2.8 7.5 342 47.50% 30.4 601 0.88 0.36 0.21 0.12 -1570 -31 -975 0% 17% 0 1
Ghost  ADC 6 83% 10 3 1.2 8.7 8.8 440 69.60% 22 446 1.2 0.51 0.27 0.24 294 -1 129 33% 0% 0
Hans sama  ADC 6 17% 2.5 2.3 2.3 3.5 8.5 378 67.10% 20 286 1.27 0.56 0.35 0.08 120 1 -306 33% 0% 0 1
JackeyLove  ADC 6 83% 9.1 3.7 1.5 10 9 441 78.20% 36.1 712 1.35 0.4 0.41 0.15 462 15 260 17% 0% 0
Kramer  ADC 6 50% 4.8 2.2 2 7.3 8.6 394 68.20% 22.7 417 1.19 0.48 0.37 0.15 -216 -8 -203 17% 17% 0
LokeN  ADC 6 67% 6.8 3.2 1.5 7 9.1 424 71.70% 20.8 410 1.14 0.39 0.37 0.18 65 9 296 17% 0% 0 1
Perkz  ADC 7 57% 4.4 4 2.6 7.3 9 431 63.90% 24.7 566 1.27 0.39 0.36 0.15 157 -4 -499 14% 14% 0
Rekkles  ADC 6 67% 7.3 1.5 1.2 7 9.4 405 65.10% 19.9 309 1.24 0.36 0.46 0.12 52 8 -102 50% 0% 0
Ruler  ADC 6 83% 5.9 4.3 1.8 6.5 9.1 460 71.30% 32.5 649 1.31 0.43 0.36 0.16 780 9 646 17% 0% 0 1
Tactical  ADC 6 50% 3.3 4 2.8 5.3 8.7 411 64.70% 27.1 472 1.12 0.44 0.39 0.14 -111 -2 244 67% 0% 0
Unified  ADC 6 33% 2.2 1.7 2 2.7 8.2 363 54.40% 27.5 388 1.1 0.49 0.29 0.2 -479 -9 -120 17% 17% 0
WildTurtle  ADC 6 50% 4.1 3.3 1.8 4.2 8.9 429 67.40% 25 443 1.07 0.33 0.3 0.12 341 0 185 33% 0% 0 3

Feel intimidated? Luckily the spreadsheets and most SQL interfaces are a lot nicer to work with than what you see above. But if this is the path you want to go down, expect to work with 10x of what you see above.

You will want to not only be good at working with these numbers but enjoy doing so. Did you enjoy mathematics in school? If not, this may not be for you. A lot of the math you learned in school may not have a lot of real-life applications, but in analytics, a lot of the content (outside of physics equations) actually plays a big role in esports analytics.

You will also want to know how you use these numbers in a visual example. Below is an example I created for my own presentation when I did a bit of esports analytical work:

As you can see, in the examples above you can see how each individual player is contributing to the team’s stats (in this case it’s team gold percentage on the left and death percentage on the right).

You’d also want to be well able with using databases. It’ll be helpful if you were fully aware of the functionalities in excel, but also SQL and any other industry-standard applications that you may need to use.

This is something you can learn, but I’d recommend at least taking an advanced course on a site like Udemy (I’ve done it, it’s well well worth it).

Creativity with Data

Are you a creative person? It’s okay if you aren’t a lot of people that work with numbers are a lot more right-side brained than left. That said, you need to be creative with data.

The thing about analytics is that big data has really taken a storm, with programs being able to capture more and more data. The limitations are less on what you can record, but rather how you can make the value of that recorded information.

Also contrasting that information is important as well. For example, the graphs below I made was a terrible use of graphics:

If you know about League of Legends esports, you’d probably be aware of how different seasonal “splits” tournaments differ. Such as the main fact that the spring season is more recreational while the summer actually requires a large amount of effort.

For that reason, obviously the spring results would be below summer matches, considering effort from both the player and the opponents being low. Not to mention the spelling error (if you didn’t notice, the first entry point should be LPL 2019 Summer Playoffs, not 2020), but that’s beside the point.

The bar chart would have been much better off comparing apples to apples, so to speak. But I failed to be creative with the massive data I had and concocted this lazy result.

The Understanding of Waste and Business Value

You’d also want to understand what actions are considered wasteful and valuable, to the team and the business. There are formulated ways you can approach certain deliverables and responsibilities you will receive, but to stand out you will need to take risks and be creative (as Team Liquid’s job posting outlined).

To do so, however, you need to understand what data collection or analysis is wasteful. This really comes from experience and understanding your craft, but if you tend to go down rabbit holes that bear no fruits, analytical work may not be the job for you. There are just so many rabbit holes you can jump into, so much data you can analyze, a lot of which provides no value to the team.

There are tools like correlation coefficients (as mentioned earlier) that can help streamline you, but that alone won’t be helpful when determining items like what specific data to collect or look at when operating amongst many different data sets.

Furthermore, you need to determine what data actually provides business value. Below is an example from a slideshow (I created, not professional top tier work) of the players in Flyquest and their stats from their summer 2020 finals season:

In this example, let’s pretend the team was evaluating all their players to determine if they want to renew their contracts 6 months from now. These pieces of information in this high-level overview slide (not the end-all-be-all, but high level) should provide information that provides the most business value.

In League of Legends’ case, since it’s a franchised league at tier 1, what may provide business value may beyond the scope of the game. It’ll likely be personality, branding and eyes. But if you are looking at who will best help you get into worlds, do these pieces of data have the strongest correlation to win-percentage?

Why did you choose these pieces of data and how does this help the administrative staff understand your conclusion on who should be resigned or not.

This qualification comes from experience, so worry not if you don’t have it, but it’s definitely good to know

Enjoyment of the Game

Lastly, one of the most important, is do you enjoy the game? You will be spending many hours during work and likely expected outside of work to be diligently watching the game and the competitive teams involved.

If you don’t enjoy watching the game, well you are going to put yourself in an unfun position. And what’s the point of working in esports if it’s not fun!

5 Steps to Becoming an Esports Analyst

So you’ve decided you want to be an esports analyst, but you don’t know where to go from here. Well, you are in luck after finding this article, below I created a high-level 5 step process to go from zero to hero in esports.

1. Starting Off as an Amateur

Let’s first determine where you are positioned. Do you have experience doing analytical work in other industries? If so, you are in a very solid position, now you need to work on your sports-specifical analytics portfolio and esports network (check out steps 3, 4 and 5).

Most of you, however, are likely in high school or college looking to penetrate into the esports industry. If that’s the case, let’s see what an esports employer is looking for in an esports analyst:

Be advised this is for a head analyst, so the experience required is a bit more extensive. However, from a quick overview we can determine a few things you’d require to secure a job.

  • A bachelor’s degree or equivalent in anything math or analytics heavy.
  • A portfolio of experience.
  • Experienced in graphical communication and strong overall communication skills.
  • The ability to use analytical tools, spreadsheets and databases.
  • Understanding of the specific game you are applying to.

For each of these points, we break them down below. To better understand where you are in the process, reflect on what you currently possess and where you need to improve.

2. Getting the Education

For those of you who are in high school, you are in luck as you can specifically select a university bachelor’s program that is proficient in your preferred field of interests. I’d recommend sports analytics if you really want to land a job in esports (more risk though as you limit your job options) or data science if you want to keep your options to different industries open.

For anyone who has a college diploma or a degree in an irrelevant field, you aren’t out of luck either. I’d recommend you look at getting a graduate certification. These programs are a one-year program for a grad cert, which is an addition to your current diploma/degree and is equivalent to a bachelor’s degree.

That said, the catch to the 1-year experience is that you must have another diploma/degree to pick it up. In addition, although it’s equivalent to a bachelor’s degree, some employers may take it with a grain of salt due to the less time in education. In that case, your portfolio (step 3) will speak volumes. As long as you have a good portfolio, you should be in a good position.

3. Building Your Portfolio

This part is likely the 2nd most important step, second only to step 4; networking (below). There are different ways to start building your portfolio as an esports analyst, but your best bet is through grassroots esports teams.

Unlike most positions for esports teams, an esports analyst is likely the least requested since it isn’t a necessity on esports teams, but therefore it’s likely the easiest position to acquire since there are many positions on the grassroots scene.

A good place to look for direct job positions is Hit Marker. That said, they have almost no esports analyst positions (at least at the time of this writing), which makes sense since again it isn’t requested much.

A very simple way to find positions on the grassroots side is by asking.

The best part about this whole ordeal is that many top-tier esports teams would likely be willing to take you on for free as well. Teams that compete in tier 1 but isn’t a Liquid or a G2, but maybe an NRG, Lazarus Esports or Luminosity Gaming.

You just need to ask. Moreover, you’d want to have a bit of a portfolio already and the ability to show your competency and seriousness (such as being in an analytical or math-focused college program).

I’d be almost completely certain Lazarus Esports would take on free team analysts. I can say this since I was heavily involved in the business side of the organization, but without a doubt, I’d say a massive number of these tier 1-competing teams (not to be confused with tier 1 teams, rather I mean teams that play in tier 1 competition) would be very open to taking on an intern analyst assuming it’s for a smaller game they are involved in and you can prove you provide some value.

This leads perfectly into the next step, which will make or break your chances of getting a job in esports.

4. Building Your Network

Networking in all industries is important, but in esports it’s vital to the success of anyone attempting to secure themselves in the industry. I’ve spoken a lot about networking on this blog, so I’m not going to go extremely in-depth on this, but I will give you some big tips to help those who haven’t read about it on this blog yet.

A very powerful way to expand your network is on LinkedIn. I’ve had a lot of success building my esports network from zero just on LinkedIn. You’d also want to attend networking events.

By volunteering or interning for organizations, you are also expanding your network. Be sure to build personal connections with these individuals, that’s one major mistake I made early in my career was not building a strong relationship with people (something I still struggle with from time to time).

5. Actively Hunting for Opportunities

Lastly, with everything setup in place, from your portfolio, education and network, now you want to hunt for opportunities. The obvious approach is to go to job boards on sites like Hit Marker and Indeed, but you also want to look at team-hire sites specifically. When you apply on those, this communicates to the team that you specifically are interested in working with that company versus if you applied on a general site trying to get as many resumes out as possible.

Also, you want it to be known to your network that you are looking for work. Once you have a decent to a strong relationship with someone, you can slide it out there that you are looking for work. They may not have a position open at their company, but they may have a relationship that is looking or they may have a position open up in the future. Regardless, having a referral to a job application is a very strong asset.