The calm before the storm
Right now some teams are preparing to play in qualifiers for four different events over the next 3 weeks. In just over a month, the first of 22 announced Majors/Minors will begin (and more are still yet to be announced). The calendar is packed — some teams will be travelling from one event to another without any breaks.
How is this any different to past years? 2016/2017 had 19 comparable events as per Liquipedia (although there were some overlapping events), so you’d think that it’s not that much more. The thing is that with the introduction of ‘qualifying points’ almost every event is worth fighting to get into, not just the Valve Majors and a few tier 1 events. Top teams (and even more for tier 1.5/2 teams) will be less willing and/or able to pick and choose which events to skip on and which to attend— it’s forcing a situation where there is huge competition for every single qualifying spot, every dollar, and every TI8 qualifying point. Since all 6 regions must have their own qualifier there will also be fewer direct invites.
Right now doing multiple concurrent qualifiers seems fine, but once teams get on the road — scheduling qualifiers around them all is going to be very difficult, and time will be very limited. My proposed solution to this is doing cascading tiered qualifiers for all the events which aren’t aiming to also to highlight the qualifiers (Dreamhack is the only organizer who I’m aware will be putting a lot of effort into the qualifiers themselves).
Cascading tiered qualifiers work by dividing teams into different tiers for each regional qualifier, and having the better teams start participating later in the qualifiers. This is slightly different to just regular tiered qualifiers [single elimination with partially filled in brackets] which have been occasionally (read: very rarely) used before (ESL Genting’s qualifiers shown below are the most recent) in that your path is not predetermined (as in, once the initial draw was done Prodota *had* to beat OG to qualify).
At first you’d separate the teams into 3 tiers (for the purposes of this example, the parameters can always be adjusted). This can be done by some objective measure (rating systems from, for example datdota / Gosugamers / Joindota), or using a power ranking by some select committee. Whatever the option used, it should be publicly announced for the interests of transparency — teams should know why they’re not invited to a specific tier. For our example let’s have the following:
- 4 teams in Tier One
- 4 teams in Tier Two
- 6 teams in Tier Three
All other teams can qualify through two open qualifiers (single elimination bo1 gauntlets). For each of the 4 rounds, a Double Elimination bracket is held, initial seedings either randomized or based on most recent placing (so a team that wins Round #1 Double Elimination would be the 5th seed going into Round #2). In Round #1 and #2,
What are the benefits of such a seemingly complex system?
Well the most obvious one is that the top teams don’t have to play that many games — a Tier One team only needs to win only 4x bo3 to qualify, whereas an Open Qualifier team needs to win 12. This means that sometimes the Open Qualifiers and first two Rounds can be played over a period of 10 days, and Round 3 can be done over 3–4 days— perfect for teams who are only available for qualifiers in specific periodic bursts.
It’s also great practice for the lower tier teams — instead of meeting a top team very early and getting slammed and knocked out— they instead will play against teams of a similar skill level before moving up to the next round. This also has the bonus of mitigating unfair/unreasonable/unlucky draws in a single elimination qualifier.
Having a system like this also means that your certainty of a decent backup team (visa issues, teams qualifying and then declining, etc) is high. In unseeded single elimination tournament without any upsets there’s almost a 50% chance that the 2nd best team comes 3rd. For organizations that are running multiple events in a year — they could also reward a team that comes 2nd in event X by a tiered invite to event X+1’s qualifiers.
In total it’s 34 bo3 matches — 10 in Round #1, 10 in Round #2, and 14 in Round 3. This can be adjusted, for example if you invited 2/6/6 in tier #1/#2/#3 and had the Round #3 be a 4-team double elimination you’d be down to only 26 matches (and the Tier One teams only need to win 3x bo3 to get the qualifier spot invitation). Other adjustments can also be made — LB matches could be turned into bo1s before the LB-Final.
Overall I think this system is transparent and fair — it also rewards the top teams by shortcutting them through extensive and unreasonable qualifiers (and saves casters/admins casting so many one-sided games). The format is simple to follow/implement in compact stages — it’s simply three double elimination brackets.