— For Gov. Bobby Jindal, Herman Cain may be something of a hero.

Or maybe it’s Michele Bachmann. Or maybe they are anti-heroes for Jindal, and his hero is one of his rivals for the 2016 Republican presidential nomination, Rick Santorum. Jindal’s hero could even be a Democrat, Barack Obama.

What those current and former presidential candidates have in common is not some core ideological principle shared with Jindal but rather a political history that argues against the validity of opinion polls at this stage of the presidential election cycle, in terms of the polls’ ability to predict the eventual winner of the election. And a belief in the fallibility of the polls is one article of faith Jindal surely needs to cling to.

The polls have not been kind to Jindal. He consistently registers in the low single digits, both nationally and in surveys in individual states, putting him near the back of the large Republican pack. The Real Clear Politics average of four recent national polls, reported at the end of last week, ranks Jindal 14th, at 1 percent. With the exception of a couple of outliers, his numbers have barely budged, despite months of campaigning.

What could be particularly discouraging to Jindal is that his results are scarcely better in Iowa, the state that hosts the Feb. 1 caucuses that launch the nomination process. Iowa is key to Jindal’s election strategy, and he has campaigned heavily there, including a five-day swing through the state last week, with plans to return Sunday.

A Real Clear Politics average of three recent Iowa polls places Jindal 12th, at 1.7 percent. He did draw 7 percent in an Iowa poll in late July that is not included in the average, and 4 percent in another survey a couple of weeks before that. But he has not sustained those gains.

Unsurprisingly, the Jindal campaign dismisses the polls as meaningless (although it is quick to call attention to any sign of improvement, however fleeting, by Jindal in the surveys, and it vigorously publicized an internal poll showing him at 8 percent in late July).

Cue the historical examples. Cain, a former CEO of Godfather’s Pizza, topped the 2012 Republican field at 27 percent in a national survey in October 2011. Bachmann, a former Republican congresswoman from Minnesota, led in a national poll with 21 percent in July 2011. Both had faded from significance by early 2012.

On the flip side — the inspiring-examples side — Jindal need look no further than Santorum, a former U.S. senator from Pennsylvania, who registered at just 3 percent in a national poll in August 2011. Santorum would rise to victory in Iowa and 10 other states in 2012 before falling short against eventual nominee Mitt Romney.

Then there’s Obama: In the 2008 election cycle, a national poll in September 2007 showed him a full 21 percentage points behind Hillary Clinton, then a U.S. senator from New York. Obama beat Clinton to win the Democratic nomination en route to the White House, later appointing her as secretary of state; she is now leading the polls in the campaign for the 2016 Democratic nomination.

Beyond anecdotes, a couple of university professors have conducted a statistical analysis of the ability of presidential polls to forecast election outcomes. They’ve found that polls taken 10 months or more before the election are essentially worthless.

Every election is different, and there is, of course, no guarantee that Jindal will mirror Santorum or Obama: The woods of the past are full of candidates who polled low early, polled low later and never got off the mat.

One way this election cycle differs from its predecessors, in terms of early poll results, is that those results apply to the sorting of candidates in debates on national TV. Fox News featured the top 10 from the polls in its debates Aug. 6, broadcasting their appearance in prime time; Jindal was relegated to the late-afternoon debate of the also-rans, which drew one-fourth the viewing audience of the main event. CNN will use a similar sorting process for its Sept. 16 debates.

If Jindal’s mind takes a dark turn, he might want to contemplate another recent presidential survey, one of no statistical significance: a straw poll of those attending a luncheon Tuesday sponsored by the East Baton Rouge Parish Republican Party. Jindal finished tied for last, with one vote among the 39 cast.

Gregory Roberts is chief of The Advocate Washington bureau. His email address is groberts@the advocate.com, and he is on Twitter, @GregRobertsDC. For more coverage of national government and politics, follow The Advocate Politics Blog at http://blogs.the advocate.com/politicsblog.