2023 Tournament Seeding Update

January 16th, 2023

With the addition of Tyler Flynn to the WATL team, one of the first tasks was examining the seeding method used in 2022 and evaluating its effectiveness at placing throwers in their ‘correct’ seeding placement. 

While ‘correct’ may be somewhat subjective in this particular instance, what we found is a large portion of top 10 throwers (in the season) who ended up seeding outside the top 100 in that same season by bullseye hit % and several instances where throwers outside the top 300 in average were seeded higher than those with the top averages in the world. We are currently working on a system that would incorporate multiple metrics and will be developing and testing that throughout 2023 (more on that to come). As a result of data-based modeling of several seeding methods, we have decided to change how 2023-sanctioned tournaments are required to be seeded.

Effective Immediately…

We are changing the metrics used for seeding for all sanctioned WATL tournaments.

Seeding for 2023 in Hatchet and Big Axe will be based on AVERAGE instead of bullseye hit %. As a tiebreaker, we will use the thrower’s % of bullseyes/Killshots/5s COMBINED (referred to as Hit% going forward) followed by the number of Killshots, and then bullseyes.

Seeding for Duals will be based on the TEAM’S average. In the case of brand new teams, the team average will be calculated as both throwers’ combined average minus 3 (this is the average by which teams’ scores decrease in duals vs as individuals). For Example, Player A has a 58 average, and player B has a 56 Average. Team AB would seed as if they had a 54 duals average (57 – 3). 

As before, your calculated average is based on your most recently completed official season (and if no recent season, most recent Tier 1/2(2023) or Tier 3/4(2022) tournament data). *The exact order of recent data for seeding is listed in the detailed explanation linked below.

Seeding for the World Championship will follow a unique model and will not follow the same seeding format of the tiered 2023 tournaments. We will release this by the end of the month so throwers can consider how to achieve their best placements. 

There were two points of concern in switching to this average which we took into consideration in making this switch:

2023 has already started, shouldn’t this have been announced sooner? I may have changed my goals in the season.

  • Since no season has been completed, and the current seeding will be based on old data, you still have time to make your 2023 seasons count for your best seeding chance. While you cannot change what you did in the 2022 seasons, it did not make sense to use an old method for seeding earlier in the year and switch after a completed season. 
  • We are conscious of the fact that we are 1-2 weeks into the season, but we also wanted to make sure this was a data-driven decision, and compiling that and making sure it was the correct decision took precedence over getting it out a week earlier.

In switching to average, won’t those who go for kills have their seeding negatively affected? 

  • Potentially. On average, those in the top 200 who went for kills in their seasons lost 6 points on the season (dropping their average by about .214 points). 
  • Killshots are not mandatory. While you do need them for Pro League, there are two very important stats to consider in the decision to try and make Pro League:
    • The vast majority of those making Pro League have their averages IMPROVE due to their Killshot attempts, so if Killshots negatively affect your average, it may be smarter to go for points in terms of average, seed, and circuit points.
    • There are very few people in the top 200 NOT attempting kills, so this has less bearing on those seeding placements than one might think (as those you are ‘competing’ with for seed also typically involve Killshot attempts)
  • Killshots are a HUGE part of the game, and really should not be ignored. The vast majority of games at worlds are decided by the Killshot throw, and a seed that takes this into account is going to be more accurate.

You can view a more detailed breakdown of the new system and the reasonings behind the change in this document. But, in short:

  • We do realize this is not the PERFECT way to do this, just the best available at this time (as we want to be able to extensively test other models before pushing them out)
  • No other metric in the sport is based on bullseye hit percentage
    • CP is assigned by score
    • ‘Best’ league is defined by score
    • Winners are decided by score (a 61 with a 70% hit rate beats a 56 with a 90% hit rate)

If anyone has any comments or questions on the change or would like any more detail on why these decisions were made (and the process that went into making these decisions), please feel free to first read the detailed discussion linked above, then contact us at: [email protected]