To second notaPINKtruck, you're going to need at least a stat book and a good bit of time if you want to get anything useful out of it. The suggestion to use excel's linear fit is a bit misguided. In that it only means something if you have a linear relationship between two variables. If you're trying to predict winning percentage from your various scouting data, then you're not really going to get anywhere with excel. In fact, you're highly unlikely to get anywhere without some sort of guess at a relationship between all your variables. In technical terms, you need a model to fit to your data. Statistics isn't going to come up with one for you. At any rate, I'm told that
The R Project is a good free statistics package, if possibly a bit more complicated than you'd like. I think when it comes down to it, this whole endeavor is going to be a bit more complicated than, say, dumping scouting data into a program and having a winning percentage or score pop out.