You (or your alliance) must have been quite lucky to get 6.10 %. I can add a few more data points from 21/10 until today.
3* | 4* | 5* | Total | |||||
---|---|---|---|---|---|---|---|---|
Red | 5 | 11.1% | 2 | 4.4% | 0 | 0 % | 7 | 15.6% |
Green | 7 | 15.6% | 2 | 4.4% | 0 | 0 % | 9 | 20.0% |
Blue | 3 | 6.7% | 4 | 8.9% | 0 | 0 % | 7 | 15.6% |
Purple | 12 | 26.7% | 0 | 0 % | 1 | 2.2% | 13 | 28.8% |
Yellow | 7 | 15.6% | 2 | 4.4% | 0 | 0 % | 9 | 20.0% |
Total | 34 | 75.6% | 10 | 22.2% | 1 | 2.2% | 45 | 100.0% |
This adds up to a total of 209 trainings since Atlantis update.
162 3-star heroes → 77.51 %
36 4-star heroes → 17.22 %
11 5-star heroes → 5.26 %
We could open a thread exclusively for results past the Atlantis update to collect a lot more data points to see whether the odds have changed or not. In the existing TC20 results thread it’s often not clear whether the data is from before and after or only from after the introduction of season 2.
As for the TC13, I usually ran two of them (until April this year) and got one 4-star hero every three to four weeks or so–basically, the same percentage other players have reported: 3 to 5 %. Thus TC13 is nice to build a good bench of 3-star heroes with the occasional 4-star surprise, but as a F2P or C2P you have to get to TC20 as quickly as possible.
Oh, and for all who are not satisfied with the TC levels as they are now, please vote for / comment in this thread: Training Camp - REDESIGN PLEASE - #10 by DJQuixo
TC 14 - 18 are basically completely useless as well as several other levels who become obsolete until you have reached a certain level. Therefore, it would be quite useful to have a better use for completely useless levels and also to reduce the gap between TC13 and TC20 by adding something inbetween that gives somewhat better chances at 4-star heroes until after several months of expanding the base eventually the awesome TC20 is ready to start