Ekko numbers: Beyond the Win Rate

Beddlam·5/31/2015, 12:52:25 PM·9 votes·2,577 views

I am using champion.gg for stats, which means we're looking at him played by and against proficient players with no mirror matches. This post uses stats available on the morning of 5/31/2015.

We finally have some stats on Ekko, and though it is very early there are some interesting things:

He is a poor farmer: weather top. middle, or jungle his cs rates are near the bottom.

**Top or mid, he causes surrenders: **If you look at his Win Rate by game length, his highest is 0-25, then it dips way down and starts climbing until he hits end-game where he really shines.

He is an effective, but not safe, assassin: In each of his roles, his matrix looks the same: average in gold, damage taken, and damage dealt with a high kill rate and poor death, assist, and win rate. He is among the best at killing opponents, but is also one of the easiest to kill, despite the safety afforded by his ult. Low assists also suggests that he may be sub-par in a team fight.

He does slightly better top: His relative stats and build are very similar in every role, but he has a slightly higher win rate top than mid or jungle.

**Conclusions: **Of course, the data sample analyzed here is relatively small (a little under 20,000 games) but it appears that he is the kind of champ that appears to be more of a problem than he is. He is a high-reward, high-risk champ, so if you are on the receiving end of the high-points of his play he feels OP, but if you are killing him a lot you don't really notice. Players are still obviously learning him, and learning how to play against him, so these stats may change drastically in the coming week. It will be interesting to track his progress.

11 Comments

LankPants5/31/2015, 6:16:21 PM3 votes

The problem I see is in the 5-15 games stats, he has a 59.9% winrate mid-lane when people have played over 5 games on him.

What does that mean? Factor out idiots who try him for the first time in a ranked game and he's winning 60% of the time.

Lugg5/31/2015, 6:33:09 PM1 votes

And this is with people not even being good with him yet. His win percentage will easily be 60% or better by this time next month and his ban rate 100%.

Beddlam5/31/2015, 6:41:02 PM1 votes

The problem I see is in the 5-15 games stats, he has a 59.9% winrate mid-lane when people have played over 5 games on him.

This is a very good point. It is a very small sample size (only 3% of Echo players), but it could suggest that his is OP but takes some time to master. It will be interesting to see what happens as more players move into that 5-15 range and the first players move into the 15-50 game range.

Beddlam6/3/2015, 4:09:01 PM1 votes

So we're up to 50,000 Ekko games and there is certainly a disturbing trend: while the 5-15 games on Ekko group has dropped to 53%, which is a strong but healthy rate, The tiny fraction of players that are above that 15 game mark are winning at a 75% clip in the middle lane. Now that's only 2% of Ekko mid players, so a bigger sample size is needed, but it could suggest that the Ekko OP outrage might be more justified than the numbers first suggested.

The other discouraging thing about the numbers is that the % of his win rate that comes in 0-25 min is even higher than before, especially for mid Ekko. That suggests that he is causing havoc early game and forcing surrenders. For a high-scaling champ like Ekko, that seems like a bad trend.

Most of the other stats have held steady though: He is still getting a lot of kills, but dying a lot and farming poorly. His damage dealt and gold earned are pretty average. Despite the high win rate for summoners with some experience on him, his over-all win rate is still sub 50%. His mid win rate has almost caught up with his top win rate, leaving jungle Ekko slightly behind.

Desiderium6/3/2015, 4:11:36 PM1 votes

I've never thought he was stronger than he actually is. Too many unskilled players in this community, even after so many years.

Rebonack6/3/2015, 4:14:51 PM1 votes

I'm glad topics like this are created. They show the board how win-rate statistics should be used.

Yes, win rate alone isn't that helpful.

But when you can weigh it by champion mastery or game-time at win, suddenly things get FAR more useful.

And just to point out, champions with a healthy degree of mastery usually boast something like a 7-10% difference in win-rate between completely new players and highly experienced players. When the difference is more like 20% (or more) then you know something is seriously wrong.

DrNova6/3/2015, 4:30:21 PM1 votes

Ekko has a 48% winrate across all elo's, and a 60's banrate. /endthread

jDVL20eCYs5/31/2015, 2:40:21 PM1 votes

ofc this bronze troll down vote a objective post that is based on relevant numbers and facts. the boards just make me sick sometimes

xLyo6/3/2015, 4:32:17 PM1 votes

its too early to discuss this because champion requires time to be played well.