Preseason 5: Popular champion picks in each lane

Happy New Year from LeagueMath.com!

Which champions are the best players in NA playing in each lane? That is the question I started out with this week, and it turned out to be an easy question to answer with data.

So, this post will show you the most picked champions in each lane by top-ranked players in NA solo-queue. But, we can do better than that, we can look at those champion's gold earnings in different phases of the game to both compare them, and contrast the different roles in the current meta-game (meta).

I will begin with the top lane.

All of the images in this post are formatted similarly:

• The most popular champion in each lane/role is on the left. So, Riven is the most-picked top-laner in this dataset, and Jayce is the fifth most picked for that role.

• Each champion has four bars above them. They measure the gold earned per minute on average by that champion in four different phases of the game. In order, they are 0-10 minutes, 10-20 minutes, 20-30 minutes, and after 30 minutes.

• The colors of the bars are relative, with dark-gray being the least, and orange being the most. So, we can see at a glance that Riven earns a lot more gold than Rumble; especially later in the game.

Some things to note. Riven's popularity is probably not undeserved. At least in terms of gold earned, she outperforms the other popular top laners.

Rumble and Gnar are quite a bit poorer than Riven and Jayce. Neither exceeds 400 gold/min at any phase of the game. The difference here when compared with Riven and Jayce is one of the biggest disparities seen in this week's analysis.

Irelia earns less gold per minute after 30 minutes than she does between 20 and 30 minutes. That seems odd, since every champion's capacity for getting gold scales up as the game goes on.

All five of these champions earn essentially exactly the same amount of gold between 0 and 10 minutes in the long run (the dataset is large enough that these averages are taken from thousands of matched played by each champion). Sometimes, to me, it seems like the outcome of the top lane is decided in the first ten minutes. But from this data, it does not look like simply picking a specific champion will give a sizable advantage.

Here is the plot for junglers.

First, a grain of salt. The metric of gold/min may be too simple to quantitatively evaluate junglers.

Again, we see that the gold earned per minute in the first ten minutes of the game is quite close between these five. Though, it is about 10 gold/min below that of the top laners above.

The increase over the four bars for each champion could be interpreted as part of that champions scaling. looking at the plot that way, it appears that Vi and Jarvan scale at least somewhat better than Lee Sin. It's interesting that Lee Sin's gold/min is the same between 10 and 20 minutes as it is from 20 to 30 minutes. Much of Lee Sin's value is cashed out in his mobility and play-making potential. He can create a lot of pressure all over the map without necessarily getting a lot of kills, or slaying minions continuously and rapidly.

Next is the plot for solo mid laners.

This plot is similar to that of the top laners presented earlier.

Zed and Orianna scale the best out of these five in terms of late-game earning potential. It does look like Zed and Syndra earn a bit more gold in the early game than Orianna does. Orianna's last-hitting relies on her passive, which scales the damage of her basic attacks with her level. At the very beginning of the game, she is a bit harder to farm with.

Ahri (like Lee Sin) is a mobile champion that can have a big impact even without tons of gold and items. Her popularity in this dataset is no big surprise. Diamond, Master, and Challenger elo (mmr) players can make good use of her skill-shot Charm skill, and she can be a wicked assassin when armed with a Deathfire Grasp.

Katarina is the fifth most popular pick here. I like how this plot shows her late-game scaling issues. Her insane mid-game damage can be mitigated by buying some magic resistance. So, though she does well between 10 and 30 minutes, it is a bit harder for her to one-shot people in the late game.

Now, here is the plot for Marksmen (bottom lane ranged AD carries).

Marksmen are the highest earners. Looking at the scale of the y-axis, we can see that their gold/min is higher than anyone else on the team. Moreover, these five popular champions are extremely well balanced with regard to their gold/min. The difference between any of these is negligible at each phase of the game.

With that said, I like how you can see by looking at the plot that Lucian and Graves have the orangest (is that a word?) bars in the late game. This matches up with my intuition that those tho (at least out of these five) scale extremely well into the late game.

Corki, Ezreal, and Vayne are a tiny bit lower than Lucian and Graves at different points in the game, but this well-balanced array of champions leads to some extremely competitive play with many interesting strategic and tactical decisions in each match-up.

Finally, here is the plot for supports.

Thresh is popular and broke, and Morgana is picked less – but rich.

The support players in this dataset sure do like skill shots. These are five of the toughest champions to play, in my opinion. It is interesting to see that Morgana earns so much more than the others in this group.

Also interesting to note is the scaling on the y-axis. No support goes above 350 gold/min, even in the late game. This is quite a bit less than marksmen and solo laners (perhaps not a big surprise), but actually comparable to junglers. Scroll back up and look.

I wonder how much of these differences come from differing choices in gold generation items. The Frost Queen's Claim line may simply earn more than the Targon's and Ancient Coin line. Maybe a topic for a future article.

All right, then. Preseason 5 marches on. Maybe this article can give you some ideas on new champions to pick up. These champions are getting disproportionately picked by the best solo-queue players in NA, so they might be good ones to learn. If nothing else, we can see which champs have extremely high skill-caps.

Did I miss anything in the analysis?

Want to see something similar or related?

Get at me.

Shy? Then just wait around until the next league-y math-y post here at LeagueMath.com.

No unpopular champion feelings were hurt in the making of this article (I hope).

Peace.