CNN’S CRITERIA for deciding who makes the grown-ups table for the GOP debate next month doesn’t make sense. It’s the sports equivalent of declaring the Red Sox losers in the iconic, come-from-behind Game 4 of the 2004 American League Championship Series against the Yankees because they scored their runs too late. The Red Sox won that game by a score of 6-4, but using CNN’s approach of averaging the current score with the score from early innings, we see the final score after nine was Yankees 2.4, Red Sox 1.8. Even after scoring two runs in the 12th, the Sox still trailed 2.8 to 2.5. They just couldn’t make up for all of those bad scores they had in the early innings. They didn’t go on to win the World Series. They were eliminated by the Yankees. Again.This is roughly what CNN’s rules will do to GOP candidate Carly Fiorina, whose recent surge has moved her from the lower tier of Republican candidates to very clearly in the top 10. To deal with the historically large field of Republican candidates, CNN (like Fox News before) is planning to divide the candidates up into two sessions: the top 10, and the also-rans. But to determine the top 10, they are using the average of eight weeks’ worth of polling, stretching all the way back to three weeks prior to the August 6 Fox News debate.

Including the old polling data is not a good idea. Poll numbers can shift dramatically in eight weeks, like the score of a baseball game can shift in later innings (thank goodness). And it’s what the score is now that should matter. Fiorina’s poll numbers have certainly shifted enough since the first debate to put her squarely in the top 10. But by giving weight to polls taken earlier in the summer, when she was still scraping bottom with the other unknown candidates, and undercounting the current score, CNN will leave her out unless a huge number of polls favorable to Fiorina are conducted over the next two weeks. And if, in the next two weeks, another lower-tier candidate catches fire and soars to the top of the pack, he will likewise be left out.

This is essentially what Carly Fiorina is facing now. She has a good score at the moment, if you take polls as a representation of the “score,” as CNN does. One could argue (and many have) that national primary polling is not a good way of determining strength, since we do not have a national primary, and the samples in this polling tend to overestimate the size of the primary electorate. But assuming they are going to use polling, including such a broad time frame makes little sense. Fiorina simply had too many 1-2-3 innings early in the campaign, and is still working them off.

Now it’s certainly possible that debating the lower-tier candidates again will work to her benefit. It seemed to help her in the Fox debate, where she outshone the competition and stayed clear of the Trump bulldozer. Indeed, that is one of the reasons for her current stronger poll performance. But in the interest of fairness, she should be where her current poll numbers put her.

Steve Koczela is a data analyst for CommonWealth and other publications and the president of the MassINC Polling Group, a subsidiary of MassINC, the publisher of CommonWealth.

CommonWealth Voices is sponsored by The Boston Foundation.

The Boston Foundation is deeply committed to civic leadership, and essential to our work is the exchange of informed opinions.  We are proud to partner on a platform that engages such a broad range of demographic and ideological viewpoints.