Gallup looks at age polarization in the Presidential election:

080618Age1_e5t6k9%201.png


Interesting chart. But the accompanying analysis says "Barack Obama's appeal to younger voters and John McCain's support among older voters may have created a situation where the outcome will turn on the preferences of middle-aged voters -- particularly those in their 40s." You see analysis of this sort all the time, but it's all based on a mistake -- there's not a demographic electoral college where "winning" particular sub-samples of the population is the key to victory and therefore it's important to focus attention on the most evenly divided demographic groups. If John McCain persuades an Obama-supporting 25 year-old to switch to his camp, that has just as big an impact as one 45 year-old one 65 year-old or one 85 year-old.

Beyond that, if you do want to label any particular group as key (for the sake of deciding which TV shows to advertise on, for example) the reasonable approach isn't to look for closely divided groups, it's to look for groups with lots of people who haven't stated a preference on the theory that those people might be easier to persuade. Voters over sixty have a marked predilection for John McCain, but there are also a lot of undecided voters in this bloc that might be worth going after. For either campaign, who "wins" seniors is irrelevant, you just go after persuadable voters, and it's arguably among seniors where the biggest group of persuadables is.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.