Analyzing Batch Volatility and Predicting Future Results

[msticky]

I have a set of results over time that show the volatility of a batch and trying to determine state at the latter end of the process whether a batch is stronger or getting weaker?

B4 has outperformed all other batches and showing an increase in latter results which means it is slowing or producing less of a result.

The lower the result number the better.

Batch is a single process that is separate from the other processes.

T is a weekly result taken and they are stagnate because they started their process at different times due to their strength.

I’m not sure what to make of these results and is there prediction to the future from the past?
Just like some fresh eyes on this and any idea on how you may see this data..
Thanks

Please check this link for the data:
https://docs.google.com/spreadsheet/ccc?key=0Ajurt2allTaddFFUZWxTcmNYRk1lbHFTTTM4MTdfSHc&usp=sharing [Broken]

 

[mfb]

It looks like there are two groups of numbers - those from 0.2 to 1.2 and a few values up to 3. I think the datasets are too small to predict the number of outliers with a reasonable precision, and those dominate the differences in the averages.
An attempt to find any time-dependence of the values looks even less promising.