I'm relatively new myself, and knowing how to price a domain in listings and what offers to accept or reject is very important. The only thing more important, probably, is getting the domain noticed by people that might actually consider buying it. Anyway, I understand the need for appraisals.
Both Sedo and GoDaddy have an algorithm that spits out some numbers. Perhaps it's based on past sales, I'm not sure. But they're both pathetic algorithms. Partly that's me speaking as a domainer, but mainly it's my background in Electrical Engineering talking--since I specialized in writing predictive algorithms based on statistical models (adaptive filters, speech recognition, that sort of thing), and I can tell when it's been done by a lazy slob.
Estibot's does a bit better, but their prices are far off the mark too much of the time. I still use tools on Estibot, but I don't really consult their appraisals and only use it maybe to sort domains in bulk as a quick filter.
It's like this ... If I had a choice about picking a domain blindfolded from a bucket of domains that Sedo or GoDaddy or Estibot appraises at $1000+ or a bucket that they appraise at $0, I'd probably pick from the first. But would I list it for $1000 blindfolded? Or bet someone $100 that it would sell at all ever? No and no.
Early on, I paid for a bunch of hand-performed domain appraisals from a particular agency (You can PM me if you want the name). Their numbers still seem to correlate well with what I perceive their value to be. That is, I usually agree with them about the domains they judged to be low $XXX and the domains they judged to be $1XXX or mid $XXXX. But the domains I've actually sold haven't agreed with their assessments OR MY OWN ASSESSMENTS. End users can behave in bizarre ways, and some of what I regard as my least interesting domains have been picked up for $XXX or $1XXX while some of what the appraisals and I agree are mid $1XXX or mid $XXXX domains get no offers.
Here's why I gave up on paying for appraisals, though: My plan was to get 100 to 200 paid appraisals and extrapolate from those to the other domains in my portfolio. So, after I received the appraisal results, I carefully constructed an extrapolation tool. Did I succeed? Yes, I did. FAR TOO WELL! What I mean is this: When I ran the domains that were actually appraised by human beings through the extrapolation tool that I'd created, I got numbers that had a correlation of 0.89 with theirs--meaning I'd accidentally reconstructed the algorithm the people I paid were using.
I'm not trying to discredit the appraisal service. There are limits to what an appraisal algorithm or even human beings can do.
If you want to know what a domain is worth, you might try looking up past sales or current asking prices. The lowest current asking price for a similar domain is probably more relevant than the really high asking prices because that domain has probably been for sale for awhile without being bought.