Of all the data included in yesterday’s WSJ poll, the number I found to be the most intriguing was not even mentioned in the news coverage. Buried in question number 21, we find the fact that 51%–a majority of Americans–say they believe Barack Obama would be the “riskier choice for president.”

This is not the first time such a question has been asked. Two weeks ago, in a CNN poll, 57% of respondents said Obama would be a risky choice. But in that poll, 54% said the same about McCain. The difference between the two polls is significant, but it was caused by structural differences. In the CNN poll, the voters were asked separately about the two candidates and gave the obvious answer: any choice of a new president can be risky. You can never know for sure what’s going to happen in a new administration. But the new WSJ poll asked the better question–which candidate is the greater risk.

There’s an interesting parallel here between the primary process and the general elections.

In the primaries, Democrats were well aware that Obama was the riskier choice (see here) and Clinton the safer one. But, as a group, they were in a gambling mood. The key words being “As a group”: not all of them voted for Obama, and according to polls those thinking Obama was riskier were  more likely to vote for Clinton. The problem for them was that there weren’t enough thinking alike.

However, there’s a big difference between the “risky” of yesterday and that of today. Although the wording is identical, the question facing Democrats in the primary season was essentially political: who has the better chance to beat the Republican nominee? Democrats realized that Obama was “riskier,” but the Party still voted for him. Their desire for a certain candidate trumped even their desire to win the election safely.

But for today’s voter, the question of “risk” is different. It’s about substance rather than politics: who’s more likely to be a terrible President? Apparently, most voters think Obama is more likely to be a bad President.  Which to me says a lot about the shrinking gap between the candidates. Come November, we might discover that the mood for gambling decreases when the question of “risk” becomes more substantial, and the decision more consequential.