Time to take another look at this -- bumping for additional suggestions. - ed.
So far, so good. We've updated our our toolbox with your suggested links and it looks like it's coming together, though we still have some room for growth. If you're on the front page, look to the left and a little down and BLAMMO, there it is, begging for your perusal. Here's what you'll find:
Looking for large amounts of collected information, sorted for your pleasure? This is where most of us get our data from, and where you can as well. Fangraphs is great for most purposes, except for base-running, in which case Baseball Prospectus' eqBRR is somewhat superior to available options. B-Ref has the preferred Wins Above Replacement for most needs, but the interface leaves a little to be desired.
Cot's is unofficially official contract data and is unbelievably valuable for so many of our larger projects. The person that could put that data into a free database would be a very popular person. Texas Leaguer's Pitch F/X database is an invaluable source for data of the same name, but take care unless you know what you're looking for.
These represent somewhat standard processes that either produce or show the results of an application of some sabermetric concept. Look toward the reference links before getting too deep with these.
Simply put, this is probably where you should start if you're new to sabermetrics. This is where we'll put primer series, how-to series, glossaries, and information best understood before using various available tools.
Read every article on BtB and still want more? This section is the best of the best. This is the sabersphere. What happens in the halls of these blogs is what we're all talking about tomorrow. They're our friends, and they should be yours as well. The best part is that sometimes you'll find conflicting opinions and different applications of the same concept, which can really get our saberbrains really pulsing.
This is by no means complete. I know there are things I missed, but need your help remembering. Please provide links and/or justification in the comments.