I imagine the new Anti-stalling feature will help a lot with the problem of “Win by Boredom”. Now the strong player can force a server decision.
However, I can’t emphasize enough that the effect of few erroneous wins is tiny compared to the effect of only playing strong players (previous comment). Here is a simulation I ran that compares a few scenarios of a new 25k:
- normal matchmaking (only play users at your current rank)
- only playing stronger players (5k)
- normal matchmaking with 3 “boredom wins” at the start
Clearly the boredom wins have an adverse effect, but, on average, the user will still pass 20k after about 15 games. On the other hand, the 25k who only plays 5k won’t pass 20k until about game 70, and won’t get to 25k for hundreds more.
The source for this simulation is here if anyone want’s to poke holes: goratings/test.py at master · benjaminpjones/goratings · GitHub
A couple of things to know about this simulation at a high level:
- Even for 25k vs 5k, we expect an occasional upset - about 1 in 200 games
- There is a lot of variability due to randomness, especially in the 5k scenario. I account for this by running the simulation 100 times, then averaging the graphs.