Mean-Variance Analysis on Team Composition against Mewtwo
Intro:
For each team of six, I sum up the individual performance score ("Overall" from Pokebattler) as the team score against a specific Mewtwo;
And then there will be eight scores (corresponding to eight different Mewtwo) for each team. Next I calculate the mean and STDEV out of the eight scores for each team;
Finally, I put the (STDEV, mean) pair of each team in a chart. The X axis is STDEV and the Y axis is the mean.
How to use the chart:
Given the same STDEV you would like a higher mean (higher risk higher return), so the points on the top are considered optimal.
Conclusion:
1. Use T Tar and Dragonite only and include at least two T Tar in your squad.
2. The more T Tar you include, the higher the expected performance your squad but also the higher the risk. Also more stardust to burn.
Any discussion is welcomed.
Answers
I always have way more counters than the crowds I raid with, but it is going to go quick here:
My Tyranitars: TT 30; TT 30; TT 30; TT26; TT 25; TT 20; TT20
Trying to get my top 5 TT to level 30--then for 6th:
5 Houndooms level 29 to 30
4 Dragonite 35; 30; 30; 27
Bug Scizor level 28
or the level 20 TTs
***Glancing at Pokebattler with novice eyes, if it is Focus Blast I might run 3 Dragonite DT/O and 3 TT B/C. Shadowball...All TT, no matter the level
Maximizing a team against Mewtwo seems like a waste of stardust. Like with the birds it seems probable that there will be a lot of people a the raids and Mewtwo will go down with 200 secs left while using your 3rd or 4th mon.
Color matching raids is more important imo. Spending dust to get 6 lvl 30 Ttars isn't effective usage just to get 1 or 2 extra balls for damage.
Right now the only situation where you should carefully build a team of six is when soloing, the only interesting part of the meta right now.