It's not an absolute fact, but rather, more of a theory. Check out last year's WL support thread for more details.
In a nutshell, schools used to have a tool called the Multiple Acceptance Report that allowed them to see what other schools their accepted candidates held As at. Data developed over many years allowed schools to pretty accurately determine their yield based on this information, so they were able to offer As above and beyond the number of seats they needed to fill with a high level of confidence that they would not be over accepting.
The new protocols coincided with a revamping of the so-called traffic rules that also involved the elimination of the MAR. This caused adcoms to be more conservative in the number of As they issued, since they had no idea where else their acceptees held As, so they didn't know what their yield would look like. Everyone who might had an A in the past but did not get one last year was placed on the WL, along with everyone who would have been on the WL even in prior years (i.e., less As and more WLs). This caused a lot of angst on the thread last year, as people were speculating when the great wave of WL movement would occur. It never really came.
There was WL movement last year, but, if anything, it was less than in prior years (and with more people than ever before on the WL). The theory comes down to basic math -- less initial As led to less people holding fewer multiple As, which led to higher yields (candidates had fewer choices, and, thus, fewer As to turn down), less seats eventually opening up, and less WL movement. The same number of seats were ultimately filled with fewer initial As and somewhat higher yields than in years past. An apparently unintended consequence of the new protocols that, in general, benefited schools, to the detriment of applicants, who ended up having fewer choices and less freedom of movement, as early CTE deadlines at some schools forced them to withdraw from WLs they would have been allowed to remain on in prior years.
This has not yet been confirmed by published numbers, but it appears to be what happened anecdotally through reporting here on SDN.