Ok, trying once again because I apparently don’t know when to quit.

Before I get into the whys and wherefores, I did want to address Marvel’s VP of Sales sticking his foot firmly past his tonsils.
Bad. And I’d explain why, but I’ve just had my post eaten twice so I’m going to take a break before I hulk out.
What we heard was that people didn’t want any more diversity. They didn’t want female characters out there. That’s what we heard, whether we believe that or not. I don’t know that that’s really true, but that’s what we saw in sales. We saw the sales of any character that was diverse, any character that was new, our female characters, anything that was not a core Marvel character, people were turning their nose up against. That was difficult for us because we had a lot of fresh, new, exciting ideas that we were trying to get out and nothing new really worked.
As smart people on Twitter and CBR have pointed out, this is total malarkey, a classic case of cherry-picking statistics to support a pre-existing narrative (in this case, blaming poor sales of new/female/diverse books on “diversity” while not extending the same analysis to the many “core” books who sold just as badly) while carefully avoiding a comprehensive analysis that might point to more systemic problems.
Systemic problems like launching and relaunching too many titles, poor marketing practices especially given the frankly insane structure of the direct market, the loss of a lot of their writing talent to Image (which gives creators much better terms when it comes to IP rights), some really boneheaded editorial decisions, and way too many crossovers.
See the thing is, people have known since the 80s that crossovers and big events boost sales – which is why editorial keeps doing them, because like management in many industries, while they might talk about innovation they tend to gravitate to strategies that work in the past. But the thing about crossovers/events back in the day is that they were like summer blockbusters: they came around once, maybe twice, a year.

Recently, though, Marvel has been accelerating the pace of their crossover events. It hasn’t been a smooth upward curve – you get years like 2010 with 9 crossovers and then 2012 only has one, then it spikes up in 2014 to 8 and then in 2014 it’s back down to 2 – but the trend has been consistently going upwards since 2009. (My source doesn’t have the number of main and tie-in issues by event, but I’d bet that’s also going up.)
As people have pointed out, this is bad for two reasons. On a business level, this many events this quickly in succession burns out existing readers who get tired of having to fork over all this money just to keep up with continuity and it discourages potential new readers who can’t find an entry point because the status quo is getting blown up every five seconds.
On a creative level, writers and artists have been complaining about crossovers being terrible from the beginning (see the image at the top, which dates back to 1989). Writers and artists who are working on the crossover get slammed with high expectations, short (and often blown) deadlines, and editorial-mandated storylines of wildly varying quality. Writers and artists who are working on regular books have their storylines put on hold to service the crossover and then get handed a new status quo which might not work at all with the stories they were telling before. Now imagine that happening five or six times a year, and it’s no wonder that talent is jumping ship.
The two things also intersect: as quality of these events decline, people are less likely to want to buy them. As the events effect the quality of regular books, people are less likely to want to buy them, especially if the most recent event featured that character doing something horrible. And the more events there are in a given year, the less any of them have any impact or stand out – it’s just like anything else, variation is the key.