Another Forum poll, another reason to throw a brick through a window.

This one claims the mayor is as popular as ever, give or take a couple of percentage points, coasting on a 42 percent approval rating, despite being exposed as a lying scumbag. (Do we even have to preface comments about Rob Ford with the word "allegedly" anymore?)

Seems every time the mayor's in crisis, Forum's there with a timely poll, conducted free of charge, to show off the firm's "research capabilities." Who's actually paying for them is another question. And how much should we read into the fact the poll was provided "exclusively" to the Ford-friendly Sun?

The media dutifully report the results. The polls are rarely examined in detail. As if the mayor isn't high enough on the stink of his own shit now that he's the talk of the planet.

News flash: what the latest Forum poll actually says is that, among Ford's supporters - let me repeat, among Ford's supporters - "the true citizens of Ford Nation," as Forum president Lorne Bozinoff calls them, 42 per cent think he's still a viable candidate for mayor.

Read another way, that means that less than half of the mayor's backers, as opposed to the electorate at large, think he's still electable.

Only, it didn't come out that way in news reports.

The Globe reported the mayor's support as being "strong despite new reports of domestic assault."

The Star reported the scandal swirling around Ford as having "little impact."

The Sun's Joe Warmington called the Forum rating "as high as any North American politician."

All of those comments skew the truth.

Political strategists are fond of saying that it's not wise to underestimate the electorate. Which is just a nice way of saying voters can be really dumb, doubly so when it comes to municipal politics, which is why guys like Ford get elected in the first place.

But in the Ford saga you can't blame folks outside the City Hall bubble for being a little confused. In the rush of information, it's easy to get swamped by the details. The mile-a-minute revelations have not only overwhelmed, but sown doubt in the public's mind. Specious polls don't help.

To most of Ford's supporters, all the mayor has done is admitted to smoking a little crack and showing up hammered on occasion in public. You need only check the comment sections of Ford-friendly papers online or listen to talk radio to understand that.

The bigger picture, that the mayor may be an accessory to murder, and possibly a home invasion, is almost a footnote in the fog.

That's not to say that the ever-widening scandal is not chipping away at the Ford brand. There's no doubt that it is.

But then a poll like Forum's comes along to reset the narrative by leaving the misimpression that the mayor has been largely untouched by the sordid stories of debauchery - coke, prostitutes, alcoholism.

Polls are supposed to be a snapshot of public sentiment. But these days they're more often shapers of public opinion, rather than reflectors. Forum polls in particular tend to be heavily skewed to the mayor's demographic: older, suburban voters, males and the poor.

Of the 1,049 persons polled in its most recent survey, more than half are over 55 and almost 80 per cent live in the burbs. The results are supposed to be weighted "to ensure the sample reflects the actual population."

But we shouldn't confuse Ford's "approval" rating with how many would actually vote for him in an election. The latter is a completely different matter.

Scroll to page 47 of the 55-page poll and it's revealed that only 33 per cent of respondents would actually vote for Ford, despite his nine-point higher approval rating.

Also missed in the reportage: the fact the mayor's approval rating is lower than each of the three other candidates mentioned as potential rivals for mayor in 2014 - Olivia Chow (59 per cent), John Tory (52 per cent) and Karen Stintz (47 per cent).

Look a little closer and you'll find a few more contradictory results. Like this one: 30 per cent of those polled who voted for Ford last time think he should resign.

There are a few other numbers that don't quite jive with the theory the mayor's looking good for 2014.

For example, more than 60 per cent approve of council stripping the mayor of his powers, some 65 per cent in Scarborough, a hotbed of supposed Ford support. Sixty-two per cent of those surveyed say Ford should resign, period.

More telling, 60 per cent of those polled say they wouldn't vote for Ford even if he checked himself into rehab and "dealt with his substance abuse problems." The interesting wrinkle here is that almost a quarter of those actually voted for Ford in the 2010.

Bozinoff concludes about those who supported Ford in the past: "They've seen the worst there is, and they still stand by their man." That first part remains to be seen. There's still a police investigation going on. The second assumes way too much.

enzom@nowtoronto.com | @enzodimatteo