For years, it’s been an axiom among political consultants that when you’re running a tough campaign, you win by getting tough. And that means you have to sling some mud.

So-called “negative” political advertising, which focuses more on an opponent’s weaknesses than the sponsoring candidate’s strengths, is something few politicians will admit to practicing but nearly all of them do it anyway, because it wins elections.

But the events of the last few weeks have seemed to turn that conventional wisdom on its ear.

According to a May 22 L.A. Times poll, the percentage of Californians likely to vote for Al Checchi in Tuesday’s election plummeted from 22 percent to 10 percent between April and May. The pollsters said the drop came about because of Checchi’s attack ads on television, based on findings that voters believed he was running the most negative campaign of all the candidates.

Polls, it must be pointed out, don’t always call the results of an election. Checchi may yet defy the pundits and pull off a come-from-behind victory. Even so, the results of the May 22 poll seemed to poke holes in the prevailing theory about negative ads.

There is a dense concentration of consultants in L.A. County who specialize in running political campaigns, and many have been struck by what happened to Checchi in the weeks before the June 2 primary.

“I don’t think I can ever remember anything this dramatic, as a backlash against negative ads,” said Larry Levine, of Sherman Oaks-based Larry Levine & Associates.

But not so fast, says veteran political advisor Harvey Englander, senior vice president with the downtown L.A.-based Kamber Group. Englander believes that backlash is not so much aimed at the negative campaign in itself, but at the candidate behind it. “People don’t like rich guys who haven’t paid their dues,” said Englander. “People aren’t going to accept somebody going negative when he hasn’t proved himself.”

As an example, Englander pointed to failed Senate candidate Michael Huffington. Like Checchi, Huffington is rich although unlike Checchi, Huffington had held political office (as a Congressman from Santa Barbara).

In addition, the “anti-rich guy” explanation doesn’t explain why voters liked Checchi in April, and didn’t like him in May. Part of the credit for that may go to a bit of negative advertising by Democrat Gray Davis who touts “experience money can’t buy” in his TV ads.

Englander believes the turnabout came because Checchi went negative too early you can’t attack well-known opponents until you have established your own credentials, he said, which Checchi had not done.

Another explanation is offered by Carl Haglund, president of West L.A.-based Dolphin Group, which is the campaign management firm for Republican gubernatorial candidate Dan Lungren.

Haglund pointed out that Checchi’s first attack ads were directed at Harman, herself a relative unknown until this election, and that things turned south for Checchi when he set his sights on Davis.

“When he attacked Gray Davis, it became obvious that he was attacking anybody who got in his way,” Haglund said. “Democrats have known Gray Davis forever in California, and they had a fairly favorable sense about him.”

Englander believes Checchi also made a tactical error by appearing to star in his own attack ads, which featured Checchi’s image. It would have been safer, Englander said, to feature surrogates attacking Democratic opponents Harman and Davis.

Checchi’s campaign was led by Robert Shrum of the well-known political consulting firm Shrum, Devine & Donilon in Washington, D.C. Shrum didn’t return calls, nor did Checchi spokeswoman Elena Stern.

One local political advisor has yet another explanation for the drop-off in the polls: the polls themselves. Back in December, when Checchi was doing well, nearly everybody said they planned to vote because so far in advance of the election, nearly everybody does. But as it gets closer to the election and people realize how little they know about the candidates, and how little they care, they get a bit more realistic about their intentions to vote.

What this means is that more people who aren’t really going to cast ballots are part of the sample in December than May. Checchi, because of his commercials, was likely the only candidate of whom these non-voters, who tend to be less educated and more apathetic about politics than voters, had ever heard.

So his seeming early support may have been little more than an illusion of the polls, and his status among real voters may have changed little between December and May.

News Editor Dan Turner writes a weekly column on marketing for the Los Angeles Business Journal.