After a scintillating discussion in another FF thread, of risks and one tool, RiskGrades, to measure it, I finally decided to check it out for myself. While I still had a number of questions about the measurement, what it means etc, it was encouraging to see many other savvier posters use RG effectively. So, I dived headlong into the RiskGrade site this weekend. After the results came out, I understood the overall measurement and its implications, but I have many more questions.  Some trivial, some perhaps more fundamental ones. So, I thought I would ask for help on the following :

1. My portfolio's overall RG score was at 36, compared to an RG scores of 91, 89 and 110 for S&P500, DJIA and the NasDaq100 respectively. I conceptually understand that my risk (as defined by RG) is only39% of S&P500. That is my portfolio has only 38% of the short term volatility of the S&P500.
2. XOM & SJT RG scores are at 119 & 153 respectively, more than almost any MF. I am holding. That seems pretty inconsistent from my simplistic observations, i.e. these 2 stocks are probably pretty slow & steady movers (up or down). So my question is is there some extra RG weight given to individual securities because they are assumed to be inherently more risky (less diversified) than MFs? As I dug into the documentation, it seems to say that it is mostly a pure statistical methodology ... ?
3. I played with what if scenarios and found that adding GRZZX to my port has a 0 risk impact, but adding BEARX actually lowers the risk impact by 2%. But GLD does not do that at all, instead it just is a zero impact ETF. Perhaps, I don't quite understand what that column (risk impact) is trying to tell me?

 Now on to more fundamental questions (probably not directly answerable, but directional answers would do) :

• I am used to seeing beta (portfolio volatility vs a benchmark) and understand that concept quite well. RG sounds like it is something similar. If so what are the differences between RG vs StdDev and Beta and RG's final answer (38% as volatile as S&P500)?
• Is there any way to measure other fundamental factors of risk of losing money over the long term (not just ST volatility)? Of course, the fundamental security analysis is one answer, but any others?
• While we measure risk, is there some other such statistical construct to measure reward? Again valuations etc are well known, and I am not asking for that, so much as some sort of a rigorous statstical way to calculate it.

All help to this new user of the sophisticated tool would be wonderful. Thanks ..... Anil