Talk:Profitability/@comment-68.97.220.191-20110711072442/@comment-4111213-20110711225543

I can confirm that buildings never need repairs on the first collection after being built or after being improved. Hundreds of samples taken and ZERO buildings repaired. However, "simply comparing the cash collections vs repairs in the awards will show a lower number" is not true for me. With 14,0001 repairs (it stops counting after 14k) I had 194,785 collections = 0.0718792. I'm not a statistics expert, but my guess is that this difference is statistically significantly higher than 1/14 given the number of samples collected. this leads me to believe now, that the standard repair rate for buildings, which have not been recently built or improved, is actually higher than 1/14th. But, proving any of this would require hand recording hundreds of thousands of collections/repairs on only a particular subset of buildings which met the criteria. Not a task I'm interested in tackling.