TI8 Regional Qualifier Distribution

Ben Steenhuisen
5 min readJun 6, 2018

--

The Open Qualifiers for TI8 were announced yesterday by FACEIT and in the last two days, VGJ Thunder and The Final Tribe were eliminated from the Supermajor thereby ending their chances to lock in an invite or snag one in the final event of the season before TI. With only TNC & Optic still trying to secure the last direct invite slot, I felt that we’re close enough to the end of the regular DPC season to begin the discussion of how the Regional Qualifiers could possibly work in terms of overall qualification for TI8.

For all of the upcoming scenarios, I’m going to assume that no matter who wins tomorrow between TNC/Optic, they don’t get enough points to make top 8 (Optic need a 4th place finish, TNC need 2nd). Although it’s very possible that they do get it, you can easily take the outcomes of this article and swap VGJ Thunder’s China slot for whoever qualifiers (so if TNC get a direct invite by displacing VGJ Thunder, then SEA will end up handing over 1 regional slot to China to balance the simulation).

The direct invites are:

  • 4 Chinese teams (LGD, Vici, Newbee, VGJ. Thunder)
  • 1 CIS team (Virtus Pro)
  • 2 EU teams (Liquid, Secret)
  • 1 SEA team (Mineski)

Then, we expect at least one slot to be given to the winner of each region. This pushes the total regional distribution up to:

  • 5 slots: China
  • 3 slots: EU
  • 2 slots: SEA & CIS
  • 1 slot: NA & SA

14 slots down, how many more to go? Well, looking at past TIs would suggest 2 or 4 more (taking the number to 16 or 18). For the purposes of this discussion I’m just going to assume 18 slots, since that’s what it was last year (the methodology I use can be replicated to just stop at 16 if that’s what you want). Now, how do we allocate these slots among the 6 regions — since someone is inevitably going to get the short straw? Well what if we counted the frequency of each region occurring in the top 20 by Glicko 2 since the first DPC event (October 2017). It leads to the following distribution.

Raw regional distribution in team-weeks

This is a bit misleading however, since it includes the teams we’ve already invited in our 14 aforementioned slots. China’s 4 directly invited teams and a regional winner, for example, must dominate that percentage of the top spots. If we remove the 8 guaranteed teams and the best other team (as a representative for the ‘regional winner’ for that region) from the calculations, we get the following updated distribution:

Regional distribution after removing invited + region-winners (team-weeks)

This suggests a few things:

  • NA clearly deserves another slot
  • It’s very tight between CIS, China & SEA for another slot.
  • EU and SA don’t really deserve any extra slot (I’m going to remove them from the rest of the calculations because of this).

Let’s say we give NA a slot (taking their total to 2 regional slots). We then have the following distribution:

Regional distribution once NA gets +1 extra regional slot

So we give NA another slot …

Regional distribution once NA gets +2 extra regional slots

SEA gets a slot …

Regional distribution once NA gets +2 extra regional slots and SEA gets +1

And hence CIS get the final slot.

This leaves us with the following final distribution:

Allocating 18 regional slots based on team-weeks spent in the top 20 by Glicko 2.

Other Variations

Perhaps Valve might instead care more about solely DPC placings rather than the actual region strength of the uninvited teams. If that’s the case — NA and SEA are in prime position (<insert Kyle quote here about strongest regions>) to get an extra slot, with NA having 5 teams who’ve accrued DPC points including 5 finals appearances and 3 wins; and SEA having Fnatic & TNC who have got DPC points on 8 occasions, and reached the finals 3 times combined (0 wins however).

China and South America are not looking great in this metric; for South America it’s just two DPC points placements and none in the finals; China is two placements with one finalist. Slot 17 could reasonably go to NA, and slot 18 could even be an EU vs CIS playoff (which also leaves a wildcard 19th team available should someone get visa issues).

Variation on the 18 slots where EU & CIS runner-ups fight for 1 slot (instead of CIS getting 2)

The final variation is that we see 20 teams invited: 8 direct invites and then the top 2 per region (2 x 6 = 12). I think that this would be overly generous to SA who’ve been given loads of chances this season with relatively few results (paiN’s 3rd place finish at a Major most recently). Despite the clear improvement of NA teams, I think it’s still much easier to make a case for a 3rd NA team, CIS team or SEA team. Since 20 slots implies 5% per slot, we can look at the initial raw regional distribution to see what the diffence/shifts are.

Allocating 2 Regional slots per region is probably grossly unfair for NA, and good for SA.

Since the Supermajor has only a few days remaining, we’ll undoubtedly hear more from Valve soon on how they’re going to ultimately divvy up the regional slots. Good luck to NA, really!

Format

I’ve been happy with the Regional Qualifier format from the last two TIs and hope it stays (bo1 round robin with top 4 playing in a double elimination bo3 and bo5 finals; bo1 tiebreakers only for splits across 4th/5th).

- Noxville

--

--