Dota 2 BTS Pro Series 9 - Americas - Playoffs Simulation
The BTS Pro Series 9: Americas playoffs kick off tomorrow, so let's get to simulating! Which team is most likely to win it all? Keep reading to find out.
Group stage recap
Before we get to the bracket, let's take a look at what happened in the group stage, and how that lines up with my earlier predictions:
| Team Name | Predicted Wins | Actual Wins | Place |
|---|---|---|---|
| INF.UESPORTS | 10.45 | 12 | 1 |
| Lava | 8.56 | 12 | 2 |
| Thunder Predator | 9.71 | 11 | 3 |
| 4 Zoomers | 10.58 | 11 | 4 |
| Hokori | 7.74 | 9 | 5 |
| D2 Hustlers | 7.19 | 7 | 6 |
| Black N Yellow | 5.82 | 4 | 7 |
| simply TOOBASED | 6.58 | 4 | 8 |
| Arkosh Gaming | 5.37 | 2 | 9 |
This ended up being way more bimodal than expected. The best teams won more, and the worst teams lost more, than my predictions had entailed. A lot of those average expected wins for the weaker teams are just the sum of their unfavored but still possible upsets, so it's not unreasonable for them to just have fallen short.
Lava (previously xXxXx) blew away expectations, tying in number of wins with Infamous in first. They're really just the Thunder Predator squad from TI10, so we knew that they had some high potential, and it looks like they've gotten back to form after a lackluster TI performance. Thunder Predator themselves have shown that their new roster is also ready to play at a high level.
The playoffs
Arkosh, the predicted weakest team coming in, won only two of their sixteen games and are going home early. The other eight teams are seeded into a double elimination bracket, with the top four starting in the upper bracket.
I simulated the bracket 50,000 times using my model (updated for group stage data, of course), to see where each team is likely to place and who is most likely to win it all.
The bracket
Here's the bracket itself:
The simulation
Here is each of those teams' probability to win first place, as well as their median finish:
| Team Name | % 1st Place | Median Place |
|---|---|---|
| Thunder Predator | 43.57% | 2nd |
| INF.UESPORTS | 24.28% | 2nd |
| 4 Zoomers | 21.84% | 3rd |
| Lava BestPc | 9.78% | 3rd |
| D2 Hustlers | 0.24% | 5th-6th |
| Hokori | 0.23% | 5th-6th |
| simply TOOBASED | 0.05% | 7th-8th |
| Black N Yellow | 0.02% | 7th-8th |
Thunder Predator's new squad is looking pretty spicy, and their performance in the group stage has bolstered my model's opinion of them. It'll be a pretty fun first-round matchup between them and Lava, who are just TP's roster up until a couple of weeks ago. Lava did 2-0 TP in the group stage, so there's some precedent here, but the stakes are turned up this time.
INF and 4Z, also starting from the upper bracket, have a decent shot at winning as well, and in fact all four top teams were within a win of each other in the group stage.
While Lava have shown that they're better than I was initially giving them credit for, I just don't see them fighting through the other top teams to win it all. They've got pretty stacked matchups against any of the bottom four teams, but are a step behind their upper bracket companions.
Meanwhile, none of the lower-bracket teams realistically have any shot at all of getting through. They've all got sub-30% matchups against the top three teams (in a best-of-one, so the best-of-three and best-of-five matches will exacerbate that even more), and would have to pull off multiple upsets in a row to get anywhere. This is Dota, so impossible is the norm, but I wouldn't hold my breath.
Placement distributions
As always, here's each team's distribution of placements:
The top four teams are pretty obvious, with little chance of a major upset there. Among them, however, the relative ranking will be a fierce fight.
TP may be the most likely to take first, but even that's only a 44% chance, so it could easily go to one of the other teams. Second through fourth could really go to anybody. The simulation results are above, but there's quite a bit of variance in the specific outcomes.
Money and markets
As of right now, there are no outright odds available on this bracket, so there's no direct market view to compare my predictions to.
I wouldn't typically have recommended betting on outrights anyway, as the vig on these types of multi-outcome bets tends to be high enough to wipe out any potential value.
Bets on each individual match, on the other hand, are where most value comes from anyway. I'll be updating my suggested bets page daily to make sure you can keep up with every single match!
For more in-depth match analysis, discussion, and predictions, as well as quantitative betting strategy tips, subscribe to stay in the loop! You can also follow us on Twitter or Facebook for quick prediction snapshots of upcoming matches. Alacrity is your fastest path from simply gambling to beating the market. Stay ahead of the game!