A Brief History of Rocket League
Part II: Pioneers & Outliers

The Evolution of Competitive Rocket League Through the Lens of RLCS League Play

TL;DR at bottom

Previously…

The first article in this series can be found here. It goes through the general trends across seasons via the stages of motor learning.

While some of the gameplay differences between players performances (especially those from different seasons) are a result of the meta changing, some of the differences between players quantify their role within the team, individual tendencies, and areas of focus.

Here, I’ll look at the factors discussed previously and how player performances can be quantified. In particular I’ve highlighted the outliers and tried to explain trends.

The data described here can be seen in the Player Factors sheet here:

Some of these stat outliers are well-visualised in the player charts from before:

• https://twobackfromtheend.github.io/player_factors_over_time_season/
  • Speed & Boost Usage
  • Aerial Affinity
  • Dribbling & Wall Hits
  • Demos
• https://twobackfromtheend.github.io/player_factors_over_time_playstyle/ (Embedded below)
  • Attacking Positioning
  • Average Boost Level
  • Shot:Save Ratio
  • Passing & Attacking Pressure

Overall Description

Each entry in the table represents the player’s appearances in a season (whether that’s 5 games as a sub, or 35 as a starter). The rankings shown below each stat header represents a ranking when the table is sorted by that stat. With a total of 401 entries, top 100 would mean in the top quarter or all RLCS season-performances (and top 20 represents the top 5%).

Stats Breakdown I

The stats that are correlated with seasons. These generally reflect gameplay improvements and meta shifts. 

Speed & Boost Usage

Recent Leaders:

• Scrub Killa, Alpha54, Chicago all appear twice (S7 and S8) in the top 20

Early Pioneers:

• Deevo appears four times (S3, S4, S5, S8) in the top 20
• Dappur (S3, S5), Mijo (S4, S5) appear twice in the top 20
• ViolentPanda thrice came in the top 35 in S3, S4, and S7

Deevo was the early pioneer of speed, making a total of 4 appearances in the top 20. In fact, Deevo’s season 2 performance lands him in the top 40 (while the season 2’s second highest entry is Tequilaz at 57^th).

Beyond them, ViolentPanda makes early top 35 appearances in both seasons 3 & 4.

Season 1 players find themselves all missing in the top 100 (out of 401 total entries), with M1k3Rules appearing 111^th. 

On the opposite side, all but one of the slowest 20 are from seasons 1 to 3, with season 5 Snaski sneaking in as the 20^th from the bottom.

The defensive players of ayjacks and Torment find their recent season 8 performance putting them 40^th and 59^th from the bottom respectively. However, finding yourself here is no indictment of your ability to win, as 4-time (as of writing) Turbopolsa makes all 7 of his appearances in the bottom 100 (17^th S1, 23^rd S2, 99^th S4, 89^th S5, 45^th S6, 75^th S7, 97^th S8; placings from the bottom). Turbopolsa’s highest rating here is lower than Chrome in season 1 and Wonder in season 6.

Aerial Affinity

Recent Leaders:

• Leading the pack, in the top 25 more than once, are EyeIgnite (S6, S7), jstn (S5, S6, S7, S8), Alpha54 (S7, S8), JKnaps (S6, S7, S8), Bluey (S4, S8), and AyyJayy (S7, S8). 
• Scrub Killa (S7, S8), Squishy (S6, S7, S8), and FlamE (S4, S8) all have multiple entries in the top 50.

Early Pioneers:

• The only top 50 appearances from the first four seasons are MummiSnow and miztik (S3) alongside Bluey, FlamE, and ViolentPanda (S4). 
• Among them, all are in the 40-50 range apart from Bluey at 17^th.

The increase in aerials since the early seasons is extremely visible as season 1 and season 2’s top aerialers of Jacob and Deevo find themselves at 91^st and 173^rd overall.

On the other side, 10 out of the bottom 15 are entries from season 1. Season 8’s most grounded players of FreaKii and Fairy Peak! are 136^th and 150^th from the bottom, being more active in the air than Lachinio (and Ferra though I’m not sure that says anything) in Season 5 .

Dribbling & Wall Hits

Recent Leaders:

• All of the top 15 come from either S7 or S8
• Alpha54, Flakes, AztraL, Sypical, jstn, Gimmick, AyyJayy, hockser, Scrub, Arsenal, and Retals in S8
• jstn, EyeIgnite, AyyJayy, and JKnaps in S7

Early Pioneers:

• Maestro (S1) at 23^rd
• Deevo (S2) at 24^th
• Squishy (S4) at 16^th

11 of the top 15 performances where the player was most likely to find that second (and third and fourth etc.) touch come from season 8. 

While the domination of the top stats by the recent seasons 7 and 8 can be attributed to improved ball control, it might be the case that opponents are respecting these players, giving more space and only challenging the following touches. These can be interpreted as related effects however, with defenders giving more space due to a better recognition of the risks of challenging the above players too hastily.

Unlike with aerials, the other side of the spectrum is a mix of early seasons, with the top 40 being entries from season 1 to 5.

Season 8’s lowest entries are (ranking from the bottom shown in brackets) Kassio (94^th), kuxir97 (96^th – Close one!), Kaydop (101^st), and Freakii (110^th), al0t (133^rd), and Turbopolsa (149^th). The fact that 4 out of these 6 made it to the World Championships indicates that not going for dribbles and wall hits (likely going for stronger hits and more direct plays) is a viable strategy.

It is interesting that the bottom-most NA player (Turbo is still ours) from season 8 is Rizzo at 194^th (from the bottom) and the bottom-most LAN-competing NA player is AxB at 214^th (from the bottom). This reflects the higher focus on retaining players’ possession within NA.

Demos

Recent Leaders:

• 4 of the top 5 are S8 entries: Arsenal, Ferra, Scrub, and Allushin

Early Pioneers:

• Fireburner (5^th) and Jacob (7^th) in S4
• Paschy90, Fireburner (36^th in S1)

Demos have recently surged in usage, with 4 of the top 5 being season 8 entries.

While the total number of demos per game has risen significantly, it is likely that the great increase in speed across the seasons have resulted in incidental demolitions even for players who never seek them.

However, looking at the “Demos near opponent’s goal” stat filters some of this effect out and enables some early-season demolishers to show up. Fireburner has been rather consistently high under this stat, landing 6 top-50 finishes. NRG’s focus on aggressiveness in season 4 can be seen as both Fireburner and Jacob take the top 2 entries in this stat (with roughly 0.43 demos near the opponent’s goal per game). The other NA duo of Chrome (top-20 in S2, S3) and CorruptedG (top-50 in S3, S4, S6) is also revealed by this stat to be an early pioneer of aggressive demolitions (though when they were on separate teams).

On the side of low demos, many entries are from the early seasons. Torment in season 6 is 8^th from the bottom, but the lowest entry from the most recent season 8 is jstn at 40^th (due to the increase in overall speed mentioned above).

Big Pad Reliance

Recent Leaders:

• Ferra takes 5 of the bottom 10 entries (S3, S4, S5, S6, S8)
• FreaKii (S8), kuxir97 (S8), Fairy Peak! (S7, S8), Speed (S7, S8), and CorruptedG (S4, S5, S7) place in the bottom 20

Early Pioneers:

• Dappur (S3, S4, S5) takes 1^st, 12^th, and 6^th from the bottom
• Ferra (S3 onwards), FreaKii (S3), CorruptedG (S3) in the bottom 30

*Leaders taken from the bottom (negative values) as a larger value in this stat is undesirable (and the overall season-to-season trend is a decrease from one season to the next).

Sikii takes all of the top four spots with entries from seasons 1, 2, 3, and 5. This shows poor boost collection decisions, although it may not have been extremely significant in the early seasons. 

In recent seasons, with the comprehensive increase in boost collections of all types, this however becomes an important consideration to put boost stats in perspective.

While season 7’s highest placing is AyyJayy at 38^th, season 8 introduces many entries near the top. In this most recent season, ayjacks entered at 12^th, Bluey at 15^th, AyyJayy at 16^th, and Flakes at 30^th.

On the opposite end, Ferra takes 5 of the bottom 10 entries, in seasons 3, 4, 5, 6, and 8. FreaKii, kuxir97, Fairy Peak!, and Speed all placed in the bottom 20 in season 8, reflecting a great boost collection distribution.

While there are early season entries at the bottom (E.g. remkoe at 21^st from the bottom in S1), they are likely a result of the low overall boost usage of that time. Only 6 out of the bottom 30 are from seasons 5 and 6.

Stats Breakdown II

The stats that are not as strongly correlated with season. As such, these stats likely reflect player playstyles, player roles within the team, and team strategies.

Attacking Positioning

Here, the top 15 in extreme forward positioning are dominated by players in the early seasons. The only recent appearances here are Ferra from season 6, Chicago from season 8, and Gimmick from season 7.

On the opposite side, Memory is a singular outlier as the most defensively positioned player ever, with a stat of −5.8 (second lowest is −3.2), from his ill-fated outing as SetToDestroyX. 

Again however, finding yourself at this side of the spectrum is not necessarily a bad thing, as recent appearances in the bottom 20 include Torment, mist, Chausette45, and Sypical from season 8 (and Chausette45 from season 7 as well). Three out of these four players made it to LAN.

Average Boost Level

While 6 of the top 15 are appearances from season 1, context is key in interpreting these high values. While the early-season appearances are more indicative of the low overall boost usage from that era, the recent plateauing of big boost collections mean that appearances in the top 25 from the recent season 8, including Flakes, Speed, Turbopolsa, and Ferra, are all indicative of great boost management.

Finding themselves with low boost across earlier seasons, Lachinio (S1, S2, S5), Gimmick (S5, S6), and Kronovi (S3, S4, S5) likely had poor boost management.

In the recent S8, Metsanauris (10^th from the bottom), Gyro (14^th), Allushin (18^th), and Memory (22^nd) were playing on a low tank for longer than most. Metsanauris and Allushin find themselves 13^th and 8^th from the bottom among the 48 season 8 entries in Big Pad Reliance, indicating that they were pushed towards low boost levels due to their role within the team, and salvaged a lot of small pads. On the other hand, Gyro and Memory find themselves 13^th and 17^th (from the top) among the same entries, showing that they had more than sufficient big boost pads and could potentially improve on their pathing over small pads. Gyro in particular, is 40^th in terms of Speed & Boost Usage in season 8 while Memory can be seen to put his boost to good use, being 1^st in Speed & Boost Usage in the same season.

Passing & Attacking Pressure

All of the top 5 spots are taken by the more attacking and dominant season 1 entries: gReazy, Maestro, Gambit, remkoe, and Kronovi – all members of either iBP Cosmic or We Dem Girlz. Markydooda from FlipSid3 and Lachinio from iBP are not far behind in 10^th and 14^th.

It would be interesting to see if the newly expanded RLCS League Plays in the upcoming season 9 will result in higher entries for the more dominant teams, pushed on by the possible higher skill gap between the best and the worst teams in the 10-team League Play

JKnaps has top 10 entries from seasons 5, 6, and 8, with a fourth entry from season 7 in 17^th, reflecting his extremely aggressive role on an extremely aggressive team.

Chicago is 6^th in season 8, 15^th in season 7 and 26^th in season 6. Rizzo’s performance in season 8 earns him an entry at 59^th,  rounding out G2’s extremely aggressive League Play playstyle with which they found great success in the S7 World Championship.

The only other team with top 50 entries in season 8 is eUnited, with Hockser (20^th) and ayjacks (43^rd), and Dignitas, with AztraL (31^st). 

On the opposite side, all 4 players from SetToDestroyX’s unfortunate outing in season 3 find themselves in the bottom 7. Set up extremely defensively, this team failed to nab themselves a series.

In the recent season 8, while the bottom-most Torment (12^th from the bottom), gReazy (14^th), and Mognus (19^th) likely find their extreme placings a result of their teams’ poor performance, Gyro’s (17^th) and mist’s (20^th) appearances are more a result of roles and team setup.

Shot:Save Ratio

Many of the top entries are from season 1’s lopsided teams (6 in the top 20). However, recent appearances do exist, with Kaydop (2^nd), Scrub (6^th), jstn (8^th) in season 7, Arsenal (5^th), Sypical (13^th), Speed (16^th), and jstn (18^th) in season 8.

These recent entries reflect offensive player roles within teams that effectively generate opportunities for these players.

13 of the bottom 25 are entries from seasons 7 and 8. Many of these players appear alongside their teammates. These teams include the following:

• Season 7
  • Gimmick (14^th), Torment (17^th), and Squishy (35^th)
  • Chausette45 (2^nd), fruity (8^th), and Ferra (24^th)
• Season 8
  • Wonder (3^rd), Kronovi (9^th), and AyyJayy (20^th)
  • Mognus (10^th), Flakes (18^th), and gReazy (22^nd)
  • FreaKii (19^th), Kassio (33^rd), and FlamE (38^th)

These teams likely set up to invite shots and respond on the counterattack.

Notes, Interpretation, and Analysis

Many of these stats are reflective of a combination of factors; these factors range from player playstyle to player role and team setup.

Unfortunately, it renders many of these stats hard to interpret.

Many of the “easy wins” involve the usage of these stats to confirm suspicions and beliefs gained through watching the games and analysing replays. However, care must be taken to acknowledge the dangers of confirmation bias.

Among these easy wins, team aggressiveness is an easy thing to identify, with G2’s recent performances land them in 1^st, 5^th, and 7^th in terms of Attacking Positioning within season 8.

Beyond the obvious, boost management is a analysable aspect crucial to recent top level play. While Average Boost Level is an indication of good overall boost management, it should be interpreted in the context of Big Pad Reliance. Specifically, while Flakes displays great boost management, he does so while taking more of his team’s big boost pads. The same applies to Turbopolsa and Speed to a smaller extent. The standout player in terms of boost management is Ferra, who holds many of the bottom spots in Big Pad Reliance while keeping his Average Boost Level high. In conjunction with his penchant for demos, he has shaped his playstyle to generate the space for Chausette45 to start his dribbles from the back.

This team setup of fruity and Ferra as devastators interfering with defences while Chausette45 sits back waiting for his chances is in huge contrast to the way some other teams set up.

In particular, the RLCS teams that struggled (arguably uncharacteristically) this season 8 of Cloud9, G2, TSM, and Complexity, all had their player with the least license to roam as their last man. 

These players (Torment, Rizzo, Alpha54, and Mognus (and gReazy)) are the most defensively positioned on their team, and lack teammates who effectively generate space for this last man to manoeuvre.

On the other hand, mist, Chausette45, Sypical, kuxir97, and Fairy Peak! are all near the top of the “spends most time as last man” list but did relatively well this season. These players are not only generally creative and mechanically skilled players in their own right, but they (importantly!) have teammates that go for a ton of demos.

These teammates who enable the playstyles of these last men are as follows (with Demos ranking within S8 shown in brackets): Retals (12^th) and Gyro (15^th), Ferra (2^nd) and fruity (5^th), Arsenal (1^st), Speed (8^th) and al0t (13^th), and Scrub (3^rd) respectively.

The main outlier to this pattern is NRG, whose line-up from first to second to last man are GarrettG, jstn, and Turbopolsa respectively. Among them, only Turbopolsa (9^th in Demos within S8) actively goes for demos and bumps.

My theory is that while their team setup might not be ideal (see: G2 & C9 as reference for teams with similarly aggressive front 2), it plays extremely strongly into the hands of their individual skills. GarrettG is the offensive player in terms of sustaining pressure through the right plays, while jstn’s mechanical abilities unlock defenses. Turbo is Turbo.

While part of Turbopolsa’s continued ability to contribute to his teams despite his low speed (By Average Speed: 4^th slowest in S8, 2^nd slowest in S7, 4^th slowest in S6, 8^th slowest in S5, 5^th slowest in S4) can be attributed to his great boost management (in terms of boost usage: lowest in S8, 2^nd lowest in S7 & S6, and 6^th lowest in S5 & S4; in terms of Average Boost Level: 3^rd highest in S8, 7^th highest in S7) , I have not yet found ways to quantify what he does on and off the pitch to my statisfaction. 

Being one of the slowest players to grace the field while having all the grace and elegance of a flying cow should not a great player make. Yet here is Turbo.

The search continues…

TL;DR

• Really. Check out my pretty graphs if you have not done so.
  
• The above breakdown was as much a TL;DR as I could create.
• The last section above is one of the more interesting ones in my opinion.