Black by Popular Demand – Activist by Choice

Pick Order Analysis – To Secure or to Counter?

What can we learn from picks order in NA LCS? Do pros prefer to pick priority champions or counter-pick, and what lanes are priorities for picks and counter-picks?

Well, it turns out there’s a pretty strong pattern for champions selection (although it depends somewhat on whether you are blue or red side). But the basics are pretty simple. Pick up your OP picks and your safe picks early. Pick up your safe picks or counter-pick in the middle. Try to counter-pick with your last picks. It should be noted that counter-picking does not necessarily mean counter-picking your lane, but can be about compositions. Ok, so that’s obvious. But what sort of counter-picking strategy do the pros prefer? Find out, after the break!

Blue Side

1st Pick

The advantage of Blue side is simple: you have first pick, allowing you to secure one champion before the other team has a chance. There are pretty much two ways to go about this:

  1. Pick an OP champion
  2. Pick a safe champion

Of course, most common first picks are a bit of both. Most commonly, first picks are junglers, supports, or mids, by a large margin. I have also listed any champions which made up more than 20% of that role’s picks.

  • Jungle – 33 (Elise 17)
  • Support – 21 (Thresh 13)
  • Mid – 20 (Lulu 9/LeBlanc 7)
  • ADC – 5 (Sivir 3)
  • Top – 1 (Renekton 1)

Also interesting is that top teams seem to have a pretty idea of what the strongest/safest champion for a given position is, as you can see by the dominance of the top picks for each position, with most of them making up half or more of the picks for that position. In the case of mid lane, it’s simply that what the power pick was shifted. Through Week 6, we had 5 first-picked LeBlancs and 2 Lulus. From Week 6 on, we had 7 first-picked Lulus and 2 LeBlancs.

And what about win rates?

  • Elise – 70%
  • Lulu – 77%
  • Thresh – 61%
  • Olaf – 66% (4/6 is limited data)
  • LeBlanc – 57% (4/7 is limited data)
  • Annie – 75% (3/4 is limited data)
  • Vi – 66% (2/3 is limited data)
  • Sivir – 66% (2/3 is limited data)
  • Renekton – 100% (1/1 is not data)

Keep in mind that the base win rate for blue side is 60%. Even with that in mind, most of these are pretty good picks.

2nd/3rd Picks

The numbers for 2nd/3rd pick may seem confisung at first, with a huge shift to ADC:

  • ADC – 57 (Caitlyn 20/Lucian 13)
  • Support – 34 (Leona 12/Annie 9/Thresh 7)
  • Top – 31 (Shyvana 11/Renekton 7/Mundo 6/Trundle 5)
  • Jungle – 25 (Vi 11)
  • Mid – 13 (Gragas 5)

Suddenly, jungle has gone from the most picked to the second-least picked, ADC and Top are up to first and 3rd from 4th and 5th, and Mid is down to last from 3rd. However, when you include the numbers from 1st picks, the picture is much more clear

  • ADC – 62 (Caitlyn 21/Lucian 13/Sivir/12)
  • Jungle – 58 (Elise 18/Vi 14)
  • Support – 55 (Thresh 20/Annie 13/Leona 13)
  • Mid – 33 (Lulu 10/LeBlanc 9/Gragas 7)
  • Top – 32 (Shyvana 11/Renekton 8/Mundo 6)

Basically, about twice as many picks are used on ADC, Jungle, and Support than on Top/Mid in the first 3 picks. Teams are saving their 4th and 5th picks for Top/Mid. This is what determines the shift between from mid → top from 1st → 3rd pick; if a mid was going to have been picked up, 1st pick would be the place, simply because the super strong mids determine the mid-game more due to roaming potential (something only Shen does top).

Looking at win rates, however, we notice something really weird: picking top lane 2nd/3rd is the way to win.

  • Top – 77%
  • Jungle – 64%
  • ADC – 56%
  • Support – 53%
  • Mid – 46%

This is a far cry from previous seasons, when counterpicking top was considered extremely important. If I had to guess, the Doran’s Shield changes will bring more varying champions into top lane, and change this data somewhat.

But more on that later. What are the top 3rd place picks?

  • Renekton – 100%
  • Sivir – 89%
  • Shyvana – 82%
  • Mundo – 67%
  • Caitlyn – 65%
  • Vi – 64%
  • Leona – 58%
  • Thresh – 57% (4/7 is limited data)
  • Pantheon – 67% (2/3 is limited data)
  • Morgana – 67% (2/3 is limited data)
  • Elise – 100% (1/1 is no data)

This data is interesting. The Sivir pick is not a bot lane counter-pick, not to ADCs, at least. In only 2 of the 9 games had an ADC already been picked, once Lucian and once Jinx. However, in every one of those games but 1, the enemy team had already picked a support. Leona was a common pick, but there were 2 Annies, a Thresh, and a Morgana. My guess is that the teams in this position simply saw Sivir as more worth securing than a support at this point, given that they could later counter-pick bot lane with their support, anyway.

4th/5th Pick

  • Top – 48 (Shyvana 11)
  • Mid – 47 (LeBlanc 6/Gragas 5/Nidalee 5)
  • Support – 25 (Leona 8/Annie 6/Thresh 5)
  • Jungle – 22 (Eve 5/Vi 4/Kha’Zix 4)
  • ADC – 18 (Sivir 7/Lucian 4)

As predicated by the data of 2nd/3rd picks, the last 2 picks are used to pick up top and mid. As we can see, the picks were much more spread out between multiple champions. Early picks may have a clear choice, but later on, those picks are taken, and there is more room to pick a specific champion to counter.

However, the teams who saw most success were those teams who picked Mid, Support, or ADC:

  • ODC – 72%
  • Support – 68%
  • Mid – 63%
  • Jungle – 55%
  • Top – 50%

So what’s the winning strategy on blue side? Just going by the data, I’d want to first pick mid whenever possible (whenever LeBlanc or Lulu make it through the ban phase). After that, I’d want to pick up top and jungle, and save ADC and Support for last.

If a key mid weren’t available, I’d want to first pick Elise, if possible, then pick up top (preferably Renekton) and ADC (probably Sivir).

If Elise is also banned, and I didn’t want to pick up another jungler, I’d grab Thresh, then pick up a top and a jungler, and finish off with an ADC and a Mid.

Why is this strategy so strong? Picking up an early mid lets you set the tone for your team. Otherwise, a jungler is important in the early/mid game. Finally, a support is a safe first-pick that can’t really get countered, as you still have an ADC to pick.

From there, you should always pick up a safe top like Renekton, as he has no bad matchups. You should look to pick up a jungler alongside Renekton, unless you already have one, at which you should grab Sivir, who is a strong ADC that can fit a number of styles depending on the Support she’s paired with.

Of course, after that, fill in what’s left.

Red Side

Red Side doesn’t have the same ability to secure a power pick, but it gets to get 2 picks after that, as well as have the last counter. This changes the pick strategy incredibly. Instead, the final counter-pick is important, and you get to choose between dominance of power picks or dominance of counter-picks.

1st/2nd Pick

Remember how for blue side, the most common early picks were Support/Jungle/ADC, with the exception of first-picked mids? Well, since you can’t first pick mids anymore, you get about the data you’d expect:

  • Support – 45 (Thresh 21/Annie 12)
  • Jungle – 40 (Elise 15/Vi 10)
  • ADC – 31 (Caitlyn 11/Sivir 8/Lucian 7)
  • Top – 28 (Renekton 10/Shyvana 9/Mundo 6)
  • Mid – 16 (Ziggs 4/Gragas 3/LeBlanc 3)

On Blue Side, Mid was a ‘first-or-last’ thing. On Red side, it’s a ‘last’ thing.

We can note that 2nd picks are a bit more versatile, with multiple champions breaking into the top picks of each position. Now, a first — and smart — thought may be that the secondary primary picks are due to first picks preventing picks of the main champion.

However, there does not appear to be significant truth to this. We can tell by comparing the % of games in which a first-pick of one position was followed by a second pick of the same position to the % of games in which that position was second-picked in general.

  • Top: 0% vs 18%
  • Jungle: 27% vs 25%
  • Mid: 5% vs 10%
  • ADC: 0% vs 19%
  • Support: 29% vs 28%

We can actually see that the percentage of second picks which are counter-picks is lower than or within 2% of the number of second picks which are securing a priority pick.

If I had to guess, the first-pick champions are good because they reveal nothing about your team composition, while simultaneously preparing you to play against anything. Once you’re picking with any knowledge of the enemy team composition, you can pick more specifically, and thus what is strong shifts away from the generically strong (Elise/Thresh) slightly towards the specifically strong (Vi/Annie), resulting in a split between the two.

As with first-picking, the win rates of any given position being second-picked are all within one game of our base win rate of 40% (being red side).

  • Support – 42.5%
  • Jungle – 42.5%
  • Top – 40%
  • Mid – 37.5%
  • ADC – 37.5%

However, there are a few champions who — when picked 2nd — yield decent win rates.

  • Renekton – 50%
  • Elise – 53%
  • Thresh – 52%
  • LeBlanc – 66% (2/3 is limited data)
  • Lucian – 57% (4/7 is limited data)
  • Jinx – 50% (2/4 is limited data)

I’m going to ignore champions with low pick rates for now, but Renekton, Elise, and Thresh are all our first-pick champions (ok, only one Renekton pick). This suggests that what makes a good second-pick probably isn’t much different from what makes a good first-pick. This makes good sense. What makes a good first-pick is being relatively insusceptible to counters, so why would you try to counter-pick such a champion? Better to then pick one of the remaining such champions.

Lulu is an exception, with an 0-3 record picked second. To be fair, those games were against TSM, Cloud 9, and CLG, the top 3 NA teams by a significant margin.

3rd/4th Pick

3rd/4th pick is much the same: grab everything that isn’t mid.

  • ADC – 41 (Lucian 14/Jinx 9/Caitlyn 7)
  • Top – 39 (Mundo 12/Shyvana 10)
  • Support – 33 (Leona 15/Annie 7)
  • Jungle – 31 (Wukong 7/Vi 6)
  • Mid – 16 (Nidalee 4/Gragas 3)

What does this tell you? Basically, if it’s this far in the pick phase, you haven’t secured an OP mid, and you have last pick, so you may as well wait for last-pick and counter-pick.

The win rates are pretty close, but there are still slight variations:

  • ADC – 44%
  • Mid – 44%
  • Top – 41%
  • Support – 36%
  • Jungle – 36%

But going champ-by-champ, we get a slightly different picture

  • Caitlyn – 71%
  • Jinx – 56%
  • Shyvana – 50%
  • Dr. Mundo – 50%
  • Renekton – 67% (2/3 picks is limited data)

Top laners picked 4th seem to do well, while the lower win rate top comes from the random top lane picks that have been pretty unsuccessful – at least in NA – Trundle and Gragas included.

6th Pick

Did you guess we’d see a lot of mid picks? Well, here they are!

  • Mid – 47 (Ziggs 5/Nidalee 5/Orianna 5)
  • Top – 13 (Renekton 4)
  • Jungle – 9 (Vi 3/Wukong 3)
  • ADC – 8 (Ezreal 4)
  • Support – 1 (Thresh 1/Lulu 1)

The 6th pick data is about what we’d expect. Basically, mid is picked first as late as possible. Again, the win rates show little variations, hovering around 40%:

  • Support – 50% (½ is barely data)
  • Jungle – 44% (4/9 is limited data)
  • Mid – 40%
  • Top – 39%
  • ADC – 38%

I feel like this is the nature of Red side: it’s more reactive. You always are going to have counter-picking as your major advantage, with 3 counter-picks available by default. As such, the importance of powerful picks is downplayed in the potential to counter-pick more. Obviously, this is true on blue side, but blue side sets the ball rolling in 100% of games with their first pick, while red side has the choice of counter-picking or setting their own balls rolling.

Counter-Picking

You’d expect counter-picking to be a big deal, wouldn’t you? Well, sort of. After all, 50% of all picks are counter-picked, by definition: you have to pick your top/jungle/mid/adc/support either before or after they pick theirs. So what position is it important to counter-pick? You’d be surprised.

The first thing to note is that the overall win rate. Counter-picking has a 47.5% win rate. This is to be expected. After all, red side gets to counter-pick more, and red side loses more. So what are the counter-pick win rates for blue and red side?

Blue side has a 58% win rate with counters (2% less than its normal win rate), and Red side has a 38% win rate with counter-picks (2% less than its normal win rate).

But weirder is what positions benefit from counter-picks. Below are the win rates for each position, overall, for blue side, and for red side:

  • ADC – 53%/64%/43%
  • Support – 50%/59%/39%
  • Mid – 49%/65%/43%
  • Jungle – 44%/55%/36%
  • Top – 43%/52%/27%

You have to be careful in reading these numbers. Yes, mid lane gives a 49% win rate when counter-picking. But this is not because counter-picking is not good. This is because said counter-picking mid is much more likely to happen on red side, which has a lower win rate, due to the tendency to counter-pick mid with last-pick. Mid lane actually has the highest counter-pick win rate on both blue and red side. So a smarter ordering of counter-picking value would be: Mid>ADC>Support>Jungle>Top.

Tagged as: , , ,

Categorised in: Analysis, eSports, League of Legends

2 Responses »

  1. Very interesting post, I have been wondering about pick priorities like this for a while, where did you get this data from? Or did you collect it yourself?

    Like

    • I gather all of my own data on the EU, NA, and GPL scenes.

      If I ever end up with OGN/LPL data, it’ll be through a friend of mine who does a similar thing.

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: