2022 LCS Spring Ratings — Rollout the Analysis

The 2022 LCS Summer Split is almost upon us, which may make a look back at Spring long overdue. However, I was inspired earlier today by looking at the Analyst rankings from The Dive

Clipped from the above video from the Official LCS Youtube Channel

All the context is provided in the video clip, but I posted an image version for those that can’t watch as they are reading. But after seeing this — I couldn’t help but think — “some of these rankings look extremely familiar to me, where have I seen this before?”

2022 LCS Spring Split Rankings via Leaguepedia

That’s right! A lot of the Analysts had predictions that were eerily similar to Spring Regular Season. Now, there have of course been MSI-off-Season roster moves, and there should be some additional context given by the Analysts for their reasoning behind doing this — but it also seems like a very safe guess when it comes to Summer predictions. As we know — not Team Liquid, 100 Thieves nor Cloud9 went to MSI — it was Evil Geniuses. The 2022 LCS Spring Playoffs played out in a much different way than anyone expected

2022 LCS Spring Playoff Bracket via Leaguepedia

By the results of Playoffs, EG would have taken #1 and bumped everyone else down a spot — while Cloud9 and 100T swapped positions in the rankings.

That got me thinking about something I had used previously to better gauge the strength of the teams during the Split, and how you could use a “better WIN%” to predict how teams might stack up going into the Summer Split. Hop on in and join the ride as we take a look at the 2022 LCS Spring results using the Colley Matrix.

Colley’s Bias Free Matrix — What is it?

Although I won’t give you the full history on the subject, the quick TL;DR is that Wesley N. Colley designed this method as a way to adjust for strength of schedule — regardless of bias from multiple different factors. Originally used in the BCS College Football system, the method can be applied to any number of sports. It gives results in a rating format that we can use to compare teams — even across different Regions!

There’s also a great article written by Percy Jaiswal that goes into the math behind finding the rating and provides Python code to calculate the rating in it’s basic form! The TL;DR of the math can be summed up pretty much like this:

via Percy Jaiswal

So instead of just normal Wins/Total Games the method is taking a running tally of performance over a stretch of time (in our case, a Split) and accounting for a team’s performance! That’s why you can call it a “better/adjusted WIN%” — we’re looking specifically at if a team won or lost — not how it was achieved.

What I’ve done to make this a little more accurate is add in some weights to the equation — mainly side selection and patch played in order to make the method work best for League of Legends. Side weighting is based off Blue/Red WIN%, while Patches are weighted more heavily towards more recent patches. As we move on, we’ll not just see the Colley Rating, but how it compares to their placing and their normal WIN% as well.

2022 LCS Spring Regular Season Colley Ratings

These results take into account the entire Spring Split (and tiebreakers for relevant teams are in parenthesis)

1. Team Liquid

Spring Placing: 1st

Spring WIN%: 77.8%

Colley Rating: 68.8% (↓ 9.0pp)

2. 100 Thieves

Spring Placing: 3rd

Spring WIN%: 66.7%

Colley Rating: 68.1% (↑ 1.4pp)

3. Cloud9

Spring Placing: 2nd

Spring WIN%: 72.2%

Colley Rating: 63.6% (↓ 8.6pp)

4. Evil Geniuses

Spring Placing: 4th (Tiebreaker Winners)

Spring WIN%: 50.0% (52.6% w/ Tiebreakers)

Colley Rating: 54.9% (↑ 4.9pp)

5. Dignitas

Spring Placing: 7th

Spring WIN%: 44.4%

Colley Rating: 48.9% (↑ 4.5pp)

6. Golden Guardians

Spring Placing: 6th (Tiebreaker Losers)

Spring WIN%: 50.0% (47.4% w/ Tiebreakers)

Colley Rating: 48.2% (↓ 1.8pp)

7. FlyQuest

Spring Placing: 5th (Tiebreaker Runner-Ups)

Spring WIN%: 50.0% (50.0% w/ Tiebreakers)

Colley Rating: 42.1% (↓ 7.9pp)

8. CLG

Spring Placing: 8th

Spring WIN%: 33.3%

Colley Rating: 37.6% (↑ 4.3pp)

9. Immortals

Spring Placing: 10th

Spring WIN%: 27.8%

Colley Rating: 34.4% (↑ 6.6pp)

10. TSM

Spring Placing: 9th

Spring WIN%: 27.8%

Colley Rating: 33.6% (↑ 5.8pp)

(PP stands for “percentage points” — and easier way to show difference between percentages)

Weighting Info

Blue Side WIN%: 37/92 (40.2%)

Red Side WIN%: 55/92 (59.8%)

Patch Weights: 0.4 -> 1.0 in 0.2 intervals (12.2 -> 12.5)

There’s a few shake ups using the Colley Rating instead of the normal WIN% rankings for the 2022 LCS Spring. Going from top to bottom — the race between TL, 100T and C9 was even closer than it felt like during the Regular Season. Liquid was still at the top, but they weren’t as clearly dominant over their close competition. 100 Thieves leapfrogged into 2nd and was the closest team to their actual WIN% in the entire LCS — while Cloud9’s Colley Rating foreshadowed a downturn in Playoffs — dropping them to 3rd overall.

EG stayed put in 4th, but their Colley Rating showed they were on an uptick from where their Tiebreakers put them (showing signs of what would come in Playoffs?) The rest of the tiebreaker teams EG faced however fell much lower than their actual rankings. GG managed to hang in at 6th like they placed, while FLY was the big downturn as a Colley Rating of 42.1% puts them out of Playoffs. DIG won big from using Colley Rating, as they would have been the 5th seed in Playoffs.

The rest of the LCS stacks up similarly, with CLG’s surge late in Spring showing some promise for the team’s future, and IMT overtaking TSM for 9th (although tiebreaker rules put TSM in 9th for the actual LCS).

While this is a good start, this doesn’t account for the way teams played in Playoffs — which would be big for attempting to rate teams for Summer Split.

2022 LCS Spring Playoffs Colley Ratings

These results take into account the entire Spring Split and then also account for the Playoff games played. While not all teams made playoffs, it still weighs on their scores for facing tougher opponents.

1. Evil Geniuses

Regular Season Colley Rating: 54.9%

Playoff Colley Rating: 72.1% (↑ 17.2pp)

2. 100 Thieves

Regular Season Colley Rating: 68.1%

Playoff Colley Rating: 67.7% (↓ 0.4 pp)

3. Team Liquid

Regular Season Colley Rating: 68.8%

Playoff Colley Rating: 63.9% (↓ 4.9pp)

4. Cloud9

Regular Season Colley Rating: 63.6%

Playoff Colley Rating: 55.6% (↓ 8.0pp)

5. Dignitas

Regular Season Colley Rating: 48.9%

Playoff Colley Rating: 48.2% (↓ 0.7pp)

6. FlyQuest

Regular Season Colley Rating: 42.1%

Playoff Colley Rating: 42.7% (↑ 0.6pp)

7. Golden Guardians

Regular Season Colley Rating: 48.2%

Playoff Colley Rating: 41.9% (↓ 6.3pp)

8. CLG

Regular Season Colley Rating: 37.6%

Playoff Colley Rating: 38.9% (↑ 2.3pp)

9. TSM

Regular Season Colley Rating: 33.6%

Playoff Colley Rating: 35.0% (↑ 1.4pp)

10. Immortals

Regular Season Colley Rating: 34.4%

Playoff Colley Rating: 34.0% (↓ 0.4pp)

(PP stands for “percentage points” — and easier way to show difference between percentages)

Weighting Info (Playoff Games Only)

Blue Side WIN%: 15/29 (51.7%)

Red Side WIN%: 14/29 (48.3%)

Patch Weights: 0.4 -> 1.2 in 0.2 intervals (12.2 -> 12.6)

Factoring in Playoff games gives us an almost exact replica of the final standings at the end of Split Season. Evil Geniuses leap frog to first place with a 17.2pp jump to represent NA at MSI, while 100 Thieves stay almost the same, although still above what their normal WIN% was. TL stumble a little but stay 3rd, while C9 tumble 8.0pp further after their Playoff implosion. Dignitas are still sitting in 5th, only dropping 0.7pp — which shows how they performed against tougher opponents (or on Blue side).

FLY get the bump up after a single game stolen from EG in Playoffs (since EG went up so high) while GG fall in a similar way FLY looked during the Regular Season ratings. CLG stay solid at 8th, even going up a bit in rating. TSM also manages to squeak by IMT by 1.0pp to take 9th place at the end of everything in the Spring.

What this does for us — especially for teams that did not make Playoffs or bowed out early — is it adjusts for the strength of their opponents overall. It gives us a bit clearer of a picture as to how they could have stacked up against the competition if Summer just started immediately after Playoffs. This of course, is not what happens in the LCS — there is a break for MSI to happen and teams make changes. The game itself has even changed, as the durability buffs took hold post-MSI and have changed things.

2022 Summer Power Rankings

Although working with imperfect information, the Colley Rating for all of Spring lets us compare how teams could start off the Summer Split. We do not know how off season acquisitions like Maple and Jensen will change teams, nor do we know how patches will change the way teams approach the game. There is no access to scrim information, so we can not add in recently played games to the Spring Colley Ratings to get a “more up to date” number. However, we can use it to inform ourselves and have some fun!

I am not calling the following accurate rankings, but only using the Colley Rating and Spring Split standings for this analysis provides a unique perspective to an Optimus Tom/Statistical 2022 LCS Summer Power Rankings:

(+/- rankings are compared to the The Dive’s Overall Average Rankings posted above)

1. Evil Geniuses (↑ 1)

2. 100 Thieves (↑ 1)

3. Team Liquid (↓ 2)

4. Cloud9 (-)

5. Dignitas (↑ 3)

6. FlyQuest (↑ 1)

7. Golden Guardians (↓ 1)

8. CLG (↑ 2)

9. TSM (↓ 4)

10. IMT (↓ 1)

This will be a fun thing to look back at, and to track during the 2022 LCS Summer Split! Games kick off tomorrow, June 17th at 2:00PM PT with EG vs C9 (my 1 vs my 4) over on LolEsports site!

If you enjoyed this analysis, consider following me here on Medium to be updated on my latest posts! You can also tune in to my Twitter account for the most up-to-date on what I’m doing, and check out my Tableau Public Profile to enjoy consistently updated LoL Esports visualizations for free!

--

--

--

Creating & telling stats-informed stories in Esports for a decade.

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

MLUX Data and Policy Panel July 2020

Making Open Data Machine-Usable

A data table. In the second column, some rows have the value “Multi-Race” while others have the value “Multiracial.”

The Dynamic Relative Strength Index

View from the Summit: How Amway is Driving a Global Business with Data Modernization

Haixun Wang Microsoft Research Paper

Why you should not be a generalist in data science

Recognizing Handwritten Digits

Get Geography by google API

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Thomas Searfoss

Thomas Searfoss

Creating & telling stats-informed stories in Esports for a decade.

More from Medium

Nokia: Data Science Co-op Summer 2022

How should you measure the success of a machine learning model?

The contributions of data science in the fight against fraud

Data Analysis and Visualizations of Chicago Divvy Bikes Sharing