Pardon? Well really I didn't know what that word meant either, but it looks quite clever. I guess it reflects the question I have which is quite a hard one to figure out and relates to coding as it's going to be coded into a php formula and secondly coders are tend to be more mathematically based. Ok, so you have a review site where everyone has to be registered to submit votes, the system is moderated to prevent people just posting values in and you have to submit a review to be able to vote. Now this site could have 100 or 1000s of products, each product having 10-20 reviews each and with such a large system for a review site it could get quite tricky. When a product has done really well and has something like 9.7 out of 10, it tends to attract a lot of people who disagree with that and knock the rating down because they like challanging the idea if it conflicts with their own, but those who think it's a 9.7 product don't feel the need to contribute because they agree. The question is, what sort of formula would you use on a review site rating system? You could use a system that uses the mean value only, but that's probably too basic and doesn't take other external values into account such as the other products in the catergory, because that category might be filled with loads of products that are rated 8 and doesn't really help differentiate to the user which one they should go for. Could it need a bit of weighting perhaps like the imdb system? But would that ruin the reputation of your site if the rating values were too faked? I'm not really sure of the solution, I guess if I'd taken statistics to university level it would be a different case. Thanks for your input fellers.