Posted by JConte on 11/18/2010 8:01:00 PM (view original):
Posted by 2chair on 11/18/2010 7:53:00 PM (view original):
Joe, I understand you could not test every little thing, but - did you test a lousy team against a good team? These are not minor or subtle flaws. But it makes a big difference to have you listening to complaints. Overall, we want these games to make sense. I think the biggest immediate concern is allowing sufficient advantage for better talent and stamina. If you want more parity, then make recruiting good talent easier for new coaches. I would like for new coaches to take over a aim and be competitive in 2-3 seasons instead of 8-12. But we should not be shocked when slid64er beats a sim! Better teams usually win, and they should not lose just because of some aebitrary desire for parity.
Of course we testing every range of team. When I run 5000 games of a 80 average rated team against a 20 average rated team, the 80 rated team wins more than 99% of the time. I think this is reasonable. The percentage moves closer to 50% as the team average ratings move closer to 50 vs. 50 where it is 50%. The progression makes perfect sense when viewed at this macro level.
Am I concerned over the number of upsets that are showing up in these early games? Honestly, yes. I want talent to be the overriding factor and we thought we had the engine calibrated accordingly. I do not currently know why so many upsets are happening. There are many very complicated interactions going on in the simulation of each game and pinpointing where changes are needed is time not a quick process.
This is where the suggestion of a beta world has merit. One thing I have learned from years of producing consumer software products is that no amount of testing can match the variability found from the creativity of people. Focussed people will find exploits fast no matter how well you test.
I have always tried to have
open betas and our best customers get involved. When you layer on top some reward system for the most creative bug found and the best exploit exposed, you will bring a lot of those problems to the surface early.
The positive is you also get a much more engaged and loyal community as contrasted by a "jilted" community when they feel they were not involved and are paying to do the beta testing anyways.