State of T3 Siege And Its Roles - Based on Statistics
I made a tweet about analyzing my own data and my thoughts on it and I found it interesting enough to do a deep dive on the T3 scene and its roles. What I found in my own case is that when I remove an outlier of Bank, there is a provable relationship for KOS,KPR, and KD of a hard support directly affecting Plant %. And no matter the circumstances, Survival % always has a direct positive coorelation with Win %. The rest of them no matter if I remove outliers do not have a provable relationship from my data set. Ask questions in the comments below if you have questions on the whole video. Thanks and enjoy!
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
I had another T3 Team send me there stats so here is my analysis of them.
Their Hard Support followed the same trend for KD and KPR with a negative correlation, they followed the trend of a negative correlation with Win Percentage and Survival Percentage, however, they were different because when they survived they planted more often with a strong positive correlation. KOS has a minimal negative correlation with plant and win percent. Plant and Win Percent have a positive correlation with player rating. Most importantly the Win% and Entry Diff% were nearly PERFECTLY positively correlated, this wasn't intentionally found. I found this by having the wrong equation by accident. However, even afterwards, Win% and Plant% had a strong positive R^2 Value of .992 which was the strongest from any of the subjects I had prior.
This team has two flex players and it is very interesting to see what they present. One of which followed the trend of Flexes are literally spontaneous and have no pattern to recognize with r^2 values of .004. The other flex however, was different. He had a strong positive correlation with win rate in every stat BUT Entry differential which was still positive but just weak.
As for the entries, both are kind of interesting individual truths. One of them had no coorelation with anything but KD and Win Percentage(r^2 of .401) while other maintained strong coorelations with only Win Percentage and Entry Differential and Player Rating(R^2 of .316 and .268 respectively).
------------------------------------------------------------------------------------------------------------------------------------------------------------------
To be honest, I am pretty disappointed with the script I wrote cause I feel like I could've made this better, happy to share summarized data charts if people want them but I can't share the individuals/teams for their own sake.
------------------------------------------------------------------------------------------------------------------------------------------------------------------
00:00 Why I did this
1:02 Methodology
2:37 Results
3:35 Entries
4:57 Flexes
6:15 Soft Supports
7:27 Hard Supports
9:19 Specific Cases and Questions to be answered
15:54 Was this worth it at all?
------------------------------------------------------------------------------------------------------------------------------------------------------------------
Discord: / discord
Twitter: / r4cc02
Twitch: / r4cc02