Did you ever test (or observe) a very new site get a couple of links from trusted places and get into the top10 (or close) of Google?
I assume you might have observed it this way?
What confuses me a bit, and makes me remain a bit skeptical is that...people such as Jim Boykin who made that post about "sorry, but I wont do SEO for your new website!" (I thikn he said he only worked with sites that were 2 years and older) said the same thing. I fidn it hard to believe that SEOs like him noticed this sandbox effect and found it so annoying simply because they got links from well-known, easy-to-get-links kind of places...i remember a post by him where he totally obsesses on where to get a link from and that getting links from the right places was better than link baiting or something along those lines.
Do I misunderstand something about what you said? I really find it extremely hard to believe that someone like him would fail to rank a new site, because he got links from the wrong, over-used places.
really confused, now!
Several years later, after much really bad speculation by the SEO community about what caused the effect (which did not actually impact all sites -- just the majority of new sites that were not really notable), some information started seeping out of Google. They more-or-less confirmed that they had been evaluating links in a completely different way from what the SEO community had assumed.
So, typically, a new site would obtain some easy-to-get links in various well-known, frequently used places. Google apparently didn't care about those links. Instead it waited for what could be called "a better sort of links" from well-established, highly trusted sites. And it took more than just a handful of such links but not a great number of them to reach some (still undisclosed) threshold where a site could then start to rank competitively.
Since then, more years have passed and Google appears to have altered its link evaluation methods in yet more ways. Many people have noted that it is now easier to get new sites ranked in Google no matter how unnotable they may be, but occasionally someone still seems to suffer from the Sandbox Effect -- where a site cannot rank, not even for its own name -- simply because it doesn't have any (or perhaps enough) trusted links pointing to it.
Of course, many people have devoted years to building up networks of trusted sites to help launch new sites into the trusted space that has become the highly coveted first stop on the road to Google success.
In a way, things have become more complicated AND much simpler since all this came out. People are no longer obsessing over the Sandbox Effect -- they just know they have to get some value-passing links from somewhere. But whereas six years ago the SEO community wasn't really set up to create/obtain those links (either legitimately or illegitimaely), now most people have at least a few trustworthy resources to work with.
I should note that a very small number of SEOs had all along encountered few instances of this Sandbox Effect. These people had from the 1990s on been practicing good link development already, and their clients benefitted from that work, even though they were obtaining links for traffic and visibility rather than to manipulate Google's search results.