Jump to Main ContentJump to Primary Navigation
LCS

Stats Science: Spring Finals Special on ADC Damage

The EU LCS Spring Split Finals are coming up in Hamburg this weekend, and we’re about to witness the best teams in Europe fight for their chance to represent the rest in MSI. However, you might be surprised to learn that some players that have made it to the finals don't necessarily have the best stats in the League.

With every big patch, new champions emerge and priorities shift. Some are better at adapting than others. A prime example of this shift in priorities is in a role that has traditionally been known for bringing the big deeps - the marksman, or ADC. The 2017 season has seen the ADC meta shift into being long-range initiators, with utility ADCs such as Jhin, Ashe and Varus rising to power.  While their damage impact in team-fights is not as high, teams still rely heavily on their consistent damage output to win games.

Therefore, I thought this would be a good time to take a deeper dive into some ADC damage statistics. Using different metrics of damage I wanted to focus on how G2's Jesper “Zven” Svenningsen and UOL's Samuel “Samux” Fernádez Fort - the ADCs who will duke it out in the Finals - measure up.

Conventional Damage Metrics

To begin, let’s look at some useful damage statistics:

  • Damage per Minute (DMG/M) tells you how much damage a player has dealt per minute over the course of the season

  • Damage Share (DMG%) measures the percent of damage dealt when compared to the player’s team.

  • Gold Share (GOLD%) looks at the percent of gold that a player had available compared to his team

  • DMG% divided by GOLD% (DMG%/GOLD%) indicates the efficiency of a player, as more gold allows for more items and thus more damage.

We can quickly see that one ADC stands above the rest. ROCCAT’s ROCCAT’s Petter “Hjarnan” Freyschuss leads the pack in all damage metrics. Not only is he #1 in DMG/M and DMG%, his 1.33 DMG%/GOLD% indicates an efficient use of resources that is above and beyond the average player. Looking at our two Finals ADCs Zven and Samux, we find a substantial discrepancy in DMG/M and DMG%. The 1.3 percentage point (PP) higher GOLD% for Zven should be taken into consideration, but is not enough to counter this discrepancy.

1

Table 1: Conventional Metrics. All stats are from Weeks 1-12 of the EU LCS Spring 2017.

Weaknesses and Biases

Both DMG/M and DMG% have their weaknesses as a performance metric. The former does not account for damage opportunities. If UOL fights a lot more than G2, Samux should have more chances to deal damage and this will be reflected in the metric. Furthermore, winning heavily influences DMG/M. A team that is steamrolling over their opposition will have more damage and it is difficult to entangle the effect that damage has on winning and vice versa (EU ADCs DMG/M in wins: 534, in losses: 462). Fortunately for us, G2 and UOL are the two best teams with fairly similar win percentages. For this reason, we can safely ignore this bias when comparing these teams. Another biasing factor is game length. Damage is higher towards the later stages of a game, so DMG/M favors players who get to the late game more often.

DMG% on the other hand should not be influenced by damage opportunities as much since it compares a player to his teammates. If UOL fights more than G2, we would assume that all players of UOL have more chances to deal damage. Comparing damage output among players on the same team means your DMG% is not only determined by your own output, but also by your teammates’ contribution. Having Tamás “Vizicsacsi” Kiss, a Top Laner who plays a lot of high damage champions, on your team will have a negative influence on your DMG%.

One problem that plagues both metrics is champion choice. Some creatures on the Rift are better equipped at dishing out damage than others. When diving deeper into the stats below, I will not provide tables for each player on each champion. This would 1) get very messy with lots of tables and 2) runs into the problem of sample size and outliers. Looking at the player’s champion pools this Split, the differences are not too large and any bias due to champion picks should not invalidate the conclusions drawn from large discrepancies in damage output. Since we disregard champion choice in the rest of our metrics, let’s look at Damage Rating which takes into consideration Zven and Samux’s performance in their respective champion pools.

 

ADC Damage Rating

This metric (used in the evaluation of Top Laner performance from a few weeks ago) uses data from EU, NA and the LCK to compare a player’s damage output on champions to the average output over all three regions while factoring in wins and losses. For a full explanation check out our “Stats Science 101: Top Laner Performance” article.

2

Table 2: EU LCS ADC Damage Rating, compared to EU, NA and LCK data

Again, Hjarnan sticks out as the only EU ADC with a positive number. It seems that our European players perform below their counterparts in NA and LCK in terms of damage output. The negative numbers reflect that they are dealing less damage than would be expected given their champion pool. Given that this metric does not account for damage opportunities or game length, it is not quite clear if this discrepancy fully reflects a weaker performance . What it does tell us though is that Zven is outperforming Samux in DMG Rating. To take a more nuanced approach and account for damage opportunities, the next section will make use of more precise data to finalize our picture of these players.

 

Damage in Fights and Skirmishes

Before showing more stats, let’s discuss damage opportunities in more detail. It is pretty clear that teams differ in their playstyles. Some rely heavily on team-fighting, while others try to force advantages through smart rotations that aim at avoiding risky encounters and getting ahead through securing more objectives. Neither is inherently better than the other and a player should not be punished in our evaluation for not fighting as much as someone else. To fully understand the impact a player has on larger skirmishes or fights, it is therefore important to only look at instances where the player is supposed to put out damage. Luckily we have the means to do this. Using granular data on damage output and the position of every player on the map, we can construct a metric that tells us if a player is close to his opponents at any given point in time.

Since I will be focusing on encounters of multiple players, I decided to look at situations where a player was within 1500 range (maximum range of Nidalee’s Javelin Toss) of at least three opponents (3+ Opp.). If there are three or more opponents within 1500 range of an ADC, it is safe to assume that he should either be dishing out damage or running away. Not all players in the EU LCS have had the opportunity to play very long games and since damage in the later stages of the game can be tremendously high, I will only analyze damage that was dealt before the 40-minute mark.

 

Analysis of 3+ Opp. Situations

The second column in table 3 shows the DMG/M in situations where three or more opponents were close. Comparing these numbers to the overall DMG/M numbers in table 1 reveals the huge impact of these situations on the final output. On average, EU LCS ADCs deal around 44% of their total damage in these kinds of situations.

The third column tells you the percentage of Game Time the player spent in 3+ Opp. scenarios (% 3+). While this should not directly impact the DMG/M 3+ numbers, it is still interesting to see which players find themselves in these situations more often. I also included information on the average number of teammates (# Mates 3+) that were around when a player was close to at least three opponents. Sometimes you might find yourself in a 2v4 situation and dying quickly without dealing damage. We would assume pro players do not end up in such dire straits too often, but we can see that there is quite some variance in # Mates 3+. Steven ‘Hans sama’ Liv for example has a higher support from his teammates when skirmishing/fighting than the other ADCs, indicating that Misfits are entering fights with a numbers advantage more often than other teams (at least when Hans Sama is involved).

3

Table 3: DMG/M in 3+ Opp. Situations (0-40 minutes)

# Mates 3+ appears to increase, on average, the damage you deal in 3+ Opp. circumstances. This can be seen in Graph 1 where I plotted # Mates 3+ (x-axis) vs. DMG/M 3+ (y-axis) on a per game basis. Each dot represents a game by an EU LCS ADC this Split (I highlighted some standout games). The trend line indicates that more teammates in 3+ Opp. situations is associated with a higher damage output. The color and size of the bubbles represent damage dealt per teammate in proximity (this is simply the y-axis value divided by the x-axis value). The deeper the green and the larger the bubbles, the more damage dealt per teammate.

4

Graph 1: Avg. Mates 3+ vs. DMG/M 3+ Opp. (per game, only ADCs)

Zven vs Samux and Standout Performers

Looking at the damage output of Zven and Samux, G2’s star deals almost 1,000 damage per minute more in 3+ Opp. instances, dominating Samux. Further investigation shows that Zven is, on average, entering fights with 0.19 teammates less in close proximity. This is not surprising given UOL’s team fighting tendencies and G2’s macro oriented style and, according to the analysis above, should favor Samux’s damage contribution.

A quick glance at Hjarnan makes us question his numbers from table 1 to some extent. He is only dealing slightly above average damage in 3+ Opp. situations, while being fairly close to the average in # Mates 3+. He is only dealing 40.7% of his total damage in fights and skirmishes, but leads the league in damage output when less people are around. Furthermore he has the highest Ranged Poke numbers of any ADC - where Ranged Poke is damage dealt when no opponent is within 1000 Range. His Ranged Poke DMG/M is 289 (0-40 minutes), 100 above the average.

Looking at the top of table 3, Martin ‘Rekkles’ Larsson boasts impressive numbers. He has shown to be a true carry for his team, but his stats should be put into context of his recent champion picks. While the ADC pool has been similar for EU ADCs this split, Rekkles is only player that has diverted from it heavily. Picks like Twitch might bias his numbers upwards to some degree.
 

Conclusion and EU LCS Finals Outlook

Finalizing our analysis of Zven and Samux, we can say that Zven looks above average in terms of his damage performance while Samux leaves a lot to be desired. Yes, none of the metrics presented above are without flaws, but put together they paint a clear picture. Samux ranks in the bottom half across the board. Relying on his strong teammates for damage output might have worked versus weaker teams, but if UOL wants to have a chance to dethrone the Kings of Europe, Samux needs to step and and try to match Zven’s impact on the game.

 

Don't miss the 2017 EU LCS Spring Finals this Sunday 23rd April, where Zven and Samux will face off in the bot lane and put the numbers to the test. The show begins at 16:30 CEST on watch.lolesports.com.