LoL Adding Structure to the 2017 Scouting Grounds Draft: A New Approach

TrevorJTrevorJ 2017-12-04 19:36:17

On December 2nd, Cloud 9 selected Top laner Ziqing "League" Zhao, making him the first player to be formally drafted in competitive League of Legends history.

FlyQuest, the Milwaukee Bucks owned franchise, was given that 1st overall selection in a randomized draft order which was later bought by Cloud 9 before the draft began.

This order was announced on Twitter and prompted some discussion in the comments on why exactly the order was random.

Riot IceChest gave insight into how they decided to randomize the order, stating “We went through a bunch of different options, but ultimately decided the most fair way would be to do a random draw."

With four teams being removed from the league and multiple imported free agents joining teams, it left Riot without many options other than randomization. Still, this begs the question: is complete and total randomization of the draft is the most fair course of action?

I would argue that no, it isn’t. Given the circumstances and the fact it’s a low stakes draft—no team will be drafting the next Bjergsen—I agree there should be an amount of randomness to it, but within reason.

Randomization without any form of structure makes for chaos and sets a poor precedent for drafts in following years where the stakes may be raised.

While there isn’t a flawless way to structure a draft under these circumstances, I’ve come up with a system that doesn’t wholeheartedly give up an attempt at doing so.

The Player Value System

When creating a system, I took into account the problems Riot faced with new franchises and high-caliber imported players entering the league.

This had me arrive at a player value based system.

Each player is assigned a point value based on their results from the previous split, for imported players these values apply to their placing in their native region:

0 points—Attend worlds by winning your region (1st Place Summer)
1 points—Attend worlds without winning your region
2 points—Reached the playoffs, but didn’t attend worlds
3 points—Placed within the “safe” range (7th-8th)
4 points—Placed in relegation range (9th-10th), played in NACS or were a bench player

The results of ranking teams in this draft order resembles something relatively representative of each team’s strength and their need for a strong academy system:

1. FlyQuest (16 total points out of 20)
1. Golden Guardians (16 total points out of 20)
3. OpTic (15 total points out of 20)
4. 100 Thieves (13 total points out of 20)
5. Clutch City (12 total points out of 20)
6. Echo Fox (11 total points out of 20)
7. Counter Logic Gaming (10 total points out of 20)
8. Cloud 9 (7 total points out of 20)
9. Team Liquid (5 total points out of 20)
10. Team SoloMid (4 total points out of 20)

A player value based system still comes with imbalances, especially when you look to certain player vs player cases, but the point is to come up with an aggregate system that will provide a better skeleton than total randomness.

Because this player point value system isn’t perfect, there still needs to be a random element to the draft order. To accomplish this, I decided on a lottery system that splits into two different lotteries to determine the overall draft order.

Lottery A will consist of the 3 teams with the highest total points drawing for picks 1 to 3 with a total of 100 combinations.

Lottery B will consist of the 7 remaining teams drawing for picks 4 to 10 with a total of 200 combinations.

The number of combinations each team receives will be based on their overall total points. Using a random number generator I tested multiple examples of a mock order and here is one of the results:

Lottery A:
1. Guardians (36% of combinations)
2. OpTic (28% of combinations)
3. Flyquest (36% of combinations)

Lottery B:
4. 100 Thieves (25% of combinations)
5. Echo Fox (17.5% of combinations)
6. Clutch City (25% of combinations)
7. Cloud 9 (6% of combinations)
8. CLG (12% of combinations)
9. TSM (2.5% of combinations)
10. Team Liquid (3.5% of combinations)

If you’d like to see the resources for the system and links to create to try making your own mock draft orders you can view that on this google doc.

The First Draft Conundrum

There isn’t a system that could avoid having inherent flaws which contradict the actual perceived strength of many teams. Removing four organizations—three of them being playoff calibre teams—makes for a system that can’t be properly weighted at the middle of the table. Factor in the blockbuster roster swaps and imported free agents and there is bound to be chaos if you try to structure teams together.

In fact, this is the crux of the issue: many of the current teams are question marks with only on paper analysis to estimate their performance. Without a basis in reality for the ranking of these newly assembled rosters, you can never come up with a system that ranks those teams with pinpoint accuracy.

For that reason, I don’t fault Riot for using a completely randomized system because it doesn’t make them beholden to any flaws an attempt at committing to a draft system would’ve.

However, the chaos in this random system gives an advantage to organizations with strong rosters that want a top 5 pick.

Two teams, Echo Fox (2nd pick) and Team Liquid (5th pick), that might’ve had a vested interest in purchasing one of those slots got them for free because of the randomization.

If Clutch City had been given better weighted odds in a structured system based on their roster they could’ve avoided paying for their #4 pick or even profited off trading it for a lower pick plus capital.

In my mock system, on average the teams that are most in need of either a) players for a farm system or b) funds to buy up players (by trading their pick) will be getting that option.

Laying a foundation with the first draft by having an aggregate order and lottery system in place that can be tinkered with in the future could’ve given more clarity and offered more hype to the draft itself.

In the future, selecting pick orders through some a type of lottery could help with getting fans involved and reinforcing the idea measures are being taken to make for a competitive league from academy teams on upward.

Interactive events like Scouting Grounds and the draft are wonderful to help develop a following for the academy scene and I hope Riot pushes to make it even more exciting in the years to come.

If you enjoyed this article or obsessive theorycrafting about League of Legends, you can follow the author on Twitter @lolTJae.

Sources: lolesports flickr, lolesports, gamepedia

There are currently no comments. Why not be the first to make one?

Just Visiting
Just Visiting