The truth about Towers since patch 6.15

https://i.ytimg.com/vi/DlDUoLXoIaI/maxresdefault.jpg This topic is popular about once a week every month, and it's no wonder why. Many long-time players who are super passionate about the game miss the old-school 50 minute games which were a test of grit determination, as I often do. With most games lasting up to only 25 minutes, there is no doubt that the game has been changed for a much faster paced playstyle that doesn't feel good for either team or the game. You either stomp, so you are never tested and cannot get better, or you get stomped, so its near impossible to attempt a comeback. The games are too long to be quick like Overwatch, and too short to be fully immersive like an RTS. Many people who post about this simply say "My game was short and unfun. Buff towers" without any room for discussion, as a different viewpoint is often shunned into the downvote abyss. It's rare these days to ever see any Rioters comment on such topics, and even harder to find one with a positive upvote count.

With this in mind, it is no wonder to see how many players think that the start of the snowball the is now an avalanche. But what if it turned out that this was not the case? What would happen if players knew that Tower first-blood did not have as big of an impact on which team won the game? Now some people who wish to uproot this discussion with their own agenda may take the hypothetical in my last out of context, so let me be clear. Tower first-blood does increase the snowball effect on the game, but it has only a small affect on which team wins.

To support my proposition, I will present the boards with a series of inforgrpahics from https://www.leagueofgraphs.com/infographics. Note that this information is taken from all games the first few days after a patch. If there is any discrepancy found in this information that would lead my analysis to be false, please notify my and I will edit in a reproposal.

Note for my understanding my analysis Many people, including many League websites get statistics wrong or misleading, so I will show how to use them correctly. If Nunu had a 50% winrate and went to a 55% winrate, there are many who will say that his winrate had a 5% increase. This is misleading. His winrate %(percentage) had a 5 point increase, but his winrate %(percentage) actually had a 10**%** increase (55 / 50 = 1.10, 1.10 - 1.00 = .10). Likewise if he fell from 55% to 50% winrate, he fell by 5 points and had a 9% decrease (50/55 ≈ 0.91, 0.91 - 1.00 = -.09).

Starting from the oldest infographic, patch 5.22, we get the following statistics. The ones of interest are located at the bottom. I recommend opening it in a new tab to see the full picture (pun intended). https://www.leagueofgraphs.com/img/infographics/patch-5-22.jpg?t=420422 Here we can see first tower leads to a 68.86% win rate. This number is most liely influenced by the buff that towers received in that patch. Since this is the lowest, I will include it in my analysis as well as the numbers in 6.14, the patch before the bonus gold, which is 69.84%. https://www.leagueofgraphs.com/img/infographics/patch-6-14.jpg?t=420422 Now here are the numbers for 6.15. https://www.leagueofgraphs.com/img/infographics/patch-6-15.jpg?t=420422 While the winrate for First Tower at 71.9% is definitely an increase, it is nothing too far from the normal ~70%. The increase from 6.15 compared to 5.22 is only a 4.4% increase (or a +3.04 point difference) and compared to 6.14 is a 3.0% increase. For reference, Renekton had a higher point difference than the towers. 6.16 had a similar tower number, but I will show it so people will see that there was little change from 6.15 to 6.16. https://www.leagueofgraphs.com/img/infographics/patch-6-16.jpg?t=420422 Now I will use the patch roughly 1 year apart which would be 6.22. We will use this (and 7.22 at the end) to compare how first tower has affected the game. https://www.leagueofgraphs.com/img/infographics/patch-6-22.jpg?t=420422 Wow, there's a pretty clear change. At 72.79%, that's a 5.7% (3.93 point) difference. What was the cause of this change? Oddly enough, it's very likely that the change of Nexus Towers from lasers to shots was the big factor. This is more obvious when compared to 6.16: notice how all of the first X by... winrates (except for first inhibitor) have an impactful increase. Also notice, the match duration: there is a clear shift to the left in time for most games. However,when compared to 5.22, they actually have strikingly similar match durations. It's possible that this is just preseason screwing up results, but it's important to note. Moving on to 7.4. https://www.leagueofgraphs.com/img/infographics/patch-7-4.jpg?t=420422 While this number is closer to the desired ~70% than patch 6.14 or 6.22, it does not have a noticeable drop when compared to patch 7.3, which had it at 72.21%. This makes it only a .76% decrease (71.66 / 72.21 = 0.9924, 0.9924 - 1.0 = -0.0076). For those wondering, this is when Riot removed the bonus 25 gold to all other members for first tower, decreasing the total gold earned from 400 => 300. While this may have helped keeps lanes equal, it had little impact on the game as a whole. You might also notice that match duration has had a significant increase, but this happened back in 6.23. So again, this change had little impact on making games longer. https://www.leagueofgraphs.com/img/infographics/patch-7-16.jpg?t=420422 Looking at 7.16 just for context, it is obvious here that games have gotten much shorter, yet winrate for first tower remained unchanged. This is likely due to Surrender-At-15 rolling out in patch 7.10, as that is where the noticeable decrease in match duration begins. Finally, we will end with 7.22 and 7.23. https://www.leagueofgraphs.com/img/infographics/patch-7-22.jpg?t=420422 https://www.leagueofgraphs.com/img/infographics/patch-7-23.jpg?t=420422 Here again, there is not a big change in the first tower winrate, and, with the winrate at a 5.2% increase from 5.22 to 7.22, it is actually lower than it was in 6.22. Also comparing match duration from 7.22 and 7.23, you can see that there is a very slight increase, maybe due to the buff to towers, but again it could just be preseason stuff. Speaking of match duration, it is important to note that, compared to 5.22 and 6.22, patch 7.22 has on average MUCH longer games. This is also noticeable when comparing patch 7.16 to patch 6.16, so anyone that claims that games have gotten shorter within the past year may have been misled. More than likely it is because lower elo players take a longer time to finish an unstoppable victory than those above them: https://www.leagueofgraphs.com/rankings/game-durations . Also, for anyone who cries "they killed our inhibitor, we might as well quit because we can't win,boo hoo hoo..." It has been that way for a long time at around a ~90% winrate, so yes. This isn't new; stop trying to emotionally manipulate the reader.

In conclusion, While the bonus gold made in made to first tower in patch 6.15 may have affected how long a game lasts and how it makes players feel to be in a very snowball meta, it has had little effect on which team wins the game. If ten thousand games were played in patch 5.22 and 7.22, 7.22 would have 358 games where the team with first tower won which is the outrageous... 5.2% (3.58 points) higher. GASP. Also just pointing out that Mundo had a bigger winrate point increase in patch 5.22 without actually getting buffed and having to pay more for items (he only got 150 health and 10 MR from and for 200 more gold), which eventually led to a whopping 77% banrate in 5.24. Just something funny I noticed.

**TL;DR First Tower with or without the extra bonus gold has had little impact on which team wins the game. **

Also, I'm also posting this to Reddit, which is why you may see this again.