It all really depends on how you measure attendance. I'll share a little information from the world of professional theme park research with you over the next few paragraphs to illustrate that different approaches can lead to very different results.
I personally think, for instance, that raw attendace numbers are pretty meaningless to the average WDW visitor. A day with 60,000 people in the Magic Kingdom during a time when the park is open late will feel a lot different than the same number of people during a day where it closes early. This is because there will be more people in the park at the same time, and those people will have less time available to tour the park, resulting in significantly longer lines. Likewise, 60,000 people 10 years ago felt much different than 60,000 people today because the park's capacity has changed. 60,000 people will also behave very differently depending on the proportion of locals vs non-local visitors. And the list goes on. You may have never thought about it this way, but what you really perceive as "crowded" while at the parks is not the total number of visitors per day, but how many visitors are in the park at any given moment, how those visitors are distributed though the park, and what they are doing at the time (having not been trained in applied statistics I just come up with my own terms for this stuff, and I call this the "specific density").
Instead of taking raw attendance, I estimate attendance patterns primarily from my own database of standby wait times and FastPass window advance rates for a set of "keystone attractions" (throwing in some weighting factors for good measure), which I believe is a more practical approach, since the majority of WDW visitors spend most of their time wanting to ride the rides.
Looking at the PassPorter chart, it appears to to take into account some of these factors (which doesn't surprise me, since it is probably the best-researched WDW guidebook you can buy today), whereas the Unofficial Guide, for instance, appears not to. One (with which I most closely agree) will tell you for instance that October is far busier than September, most of November, and even much of August, whereas the other will have you believe that October actually has the second-lowest visitation of the year.
And so it goes....
PS: Resort attendance is a fairly poor proxy for theme park attendance, since it assumes that locals are just as likely to stay on-property as tourists.