onGamers News

Are gold leads too snowbally? An investigation into gold leads and win rates.

A gold lead as early as 10 minutes leads to an overwhelming 79% win rate in NA and Korea despite Riot's recent changes.

In Season 4 one of Riot's big goals was to decrease the amount of snowballing in League of Legends. This came through in many recent changes, such as: the reduced gold for kills before 2 minutes, the harder to kill early turrets and overall decreased incentive for 2v1ing, the lower gold from dragon, scaling assist gold, and champion balancing. This was done in order to make late game a more important phase of the game promoting the possibility of comebacks. However, looking at the data this tells a different story. Overall, win rates are up after getting a gold lead at 10 minutes, with the leader winning almost 80% of their games in North America and Korea. Despite all these early game changes, was this what Riot intended?

While the win rates go up, the actual lead goes down.

While the win rates of a 10 minute lead have increased significantly, the actual lead has decreased from on average 1,146 to 925. This means despite the actual lead going down the win rates have gone up, does this mean teams can just snowball with a smaller lead? Is a 900 gold lead at 10 minutes (given the average gold at 10 minutes in NA is 13.3k, that is a 7% gold lead) actually significant enough to determine the winner 80% of times? Instead, these changes that Riot implemented might have been pointed at a different thing, reducing the amount of "cheese" in the game, creating an environment where the "better" team gets the lead more consistently early. Without level 1's and 2v1 situations, there is less room to make one mistake which throws the game completely. Meaning teams like TSM, who have a strong and consistent lane phase, get a lead more often. However, is this more stable style of the game an improvement for the spectator experience? Knowing that there are so few comebacks from a lead as early as 10 minutes makes the game seem somewhat anti-climatic.

Team breakdown of early leads in NA

When looking at the breakdown of teams you can see how the top teams get a lead at an overwhelming manner. The biggest exception is EG who has been historically having trouble finishing out games. And despite teams like TSM and CLG getting a lead in a significant portion of their games, I wouldn't call these teams early game focused. TSM in fact averages 13.4k gold at 10 minutes, which is 5th in the league. However, what is important to them is that they don't get cheesed, they don't give up kills and this lets them ride to a consistent advantage.

Note: "NA S3" stats are from the second half of summer split S3.

Design by @SrcsmApprciated

Collection Help by: @PelkaSupaFresh, @Whedgehead, @Traepoint, @jakeFLAC, and Kathix

2 Comments
Edited by beepboop

This is pretty sensationalist. In a region where the skills of the teams are much closer, the win percentage drops significantly (looking at EU). In regions where the skill of the teams are more obvious (KR and NA) this percentage is bigger. The fact that most teams don't know how to play from behind plays a more significant role in this win percentage than snowballing a less than 1k gold lead.

Posted by Enzo

This is why TSM games are boring. They're comps hardly change, they hardly try anything really new. Whatever, though, I think that kind of strat eventually fails.