Battle Pass Predictions — TI 2022
Our predictions for the Battle Pass were made using data from https://stratz.com/battle-pass/predictions. Check it out if you want to find the best picks for yourself!
The data was sourced from all matches played at the Last Chance Qualifiers (LCQ), all Regional Qualifiers (RQ), the Arlington Major, and all of the most recent Division 1 Tours (D1 Tours). However, the data is specific to only the matches where TI-Qualified teams played, and is further specific to the players on those teams.
The filters I used, and my approach for each category, is below. There were several scenarios where two or more picks were extremely close (e.g. only 1% difference in win rate, etc). I attempted to use minimal interpretation of the data — you may prefer to choose lower then the top picks based on your preference or a different choice of filters.
Heroes Predictions
Categories: Most Picked, Most Banned, Highest Win Rate
Filter: LPQ only Link
Short Explanation: These 3 categories require only the most up-to-date meta possible.
Categories: All the other Heroes predictions
Filter: LPQ and RQs — Link
Short Explanation: I added the RQs for these to expand both the match count and team coverage. I didn’t use older leagues to prevent their staler metas from introducing greater uncertainty.
Teams & Players Predictions
Categories: All Teams & Players categories
Filter: All Leagues in dataset — Teams Link and Players Link
Short Explanation: Many of the teams and players that didn’t play in TI didn’t play in the Qualifiers. As a result, in order to get data for all the teams, I had to use the entire dataset.
Tournament Predictions
Category: Games Played at Main Event
Filter: TI 10 Main Event — Link
Short Explanation: In order to approximate the number of matches played in the Main Event, only leagues that have the exact same format can be compared. In this case, the TI 10 Main Event is the best candidate, and 50 matches were played in it.
Categories: Total Heroes Picked, Total Heroes Banned
Filter: Arlington Major (with assistance from cross-referencing TI10) — Link
Short Explanation: While the Arlington Major had 103 Heroes picked and 98 Heroes banned, leaving it just short of the 101+ banned required for my prediction, the combined Group Stage and Main Event saw almost every hero banned at least once.
Categories: All the other Tournament predictions
Filter: Arlington Major (with some cross-examination of RQ, LCQ, and D1 Tours) — Link
Short Explanation: All of the predictions in these categories involve extremes (most, longest, shortest, etc). By using the Arlington Major matches as a testing bed for what is likely to occur in terms of extremes, then double checking the other leagues to see if any outliers existed, I was able to create reasonable expectations for TI.
The one caveat is that, since there were very few (5) total instances of 15:00 to 19:59 duration matches across all leagues, and only 1 instance of a 10:00 to 14:59 duration match, I chose to pick 15:00 to 19:59 as my prediction. This is because in the only <15 minute match, RNG forfeit without losing a single tier 2 tower. I speculate that it’s unlikely that a team will do that at TI (source: https://stratz.com/matches/6695580014).
Good luck with your picks!
Enjoy keeping up with what’s happening at STRATZ? Follow us on Medium, Twitter, or Facebook to get notifications of our latest Supdates! If you want to get more involved, join us on Discord!
Interested in our most recent updates? Look back at the Battle Pass Update Part 2!
Thanks for reading!
