Today’s the day Fox News announces the finalists for Thursday’s main event. Until Kasich got in, Perry was just inside the bubble. Now he’s outside, and as Harry Enten notes, not even rounding up or down to the next percentage point will help him.

He spent the last month trying to muscle into Trump’s media spotlight by attacking “Trumpism,” but in the end, according to the Media Research Center, he’s earned just four minutes of coverage on the three broadcast networks’ evening news shows since January. (Trump himself has earned 116 minutes.) Barring any surprising new polls in the next few hours showing him surging past Kasich, the only way Perry ends up on the same stage as Trump now is if Fox abandons its format entirely and decides to hold one mega-debate with all of the candidates or if it sets a different arbitrary polling line for the main debate to include Perry — “all candidates with a polling average of two percent or greater,” say, instead of those with the 10 highest averages. But if Fox does that, they run into another problem. Namely, the more big-name candidates they rope into the 9 p.m. main debate, the less compelling the 5 p.m. consolation-prize debate becomes. A 5 p.m. debate involving Perry, Jindal, Fiorina, Santorum, and Lindsey Graham could be interesting. Take Perry out and it’s less interesting. Take Santorum out by rounding his average up to two percent and it’s less interesting still. How small do you want to make the 5 p.m. debate before everyone loses interest?

To get a sense of just how arbitrary these polling criteria are, eyeball this table from yesterday’s new Monmouth poll. Remember, given the size of polling samples and the tiny spreads between candidates like Kasich and Perry, the difference between making the main debate and missing it is a matter of fewer than 10 people in many polls. If just five respondents, say, flipped from Kasich to Perry in each poll, Perry would likely be in while Kasich is out. Lucky thing for the RNC that it played out the way it did, though, as it’s a lot easier to justify excluding a former governor of a reliably red state like Texas than the sitting governor of a must-have swing state like Ohio. Anyway, the Monmouth data:

Taking the margin of error into account, it’s possible that Perry’s true support is as high as three percent while Kasich’s is as low as 1.5 percent, and yet it’s Kasich who’s almost certainly in the main debate while Perry is out. In fact, as Monmouth notes, only five candidates — Trump, Bush, Walker, Cruz, and Huckabee — can safely be said in this poll to be in the top 10 in the field taking into account the margin of error. Conceivably, Perry outpolled even Marco Rubio(!) here. Obviously that effect is reduced by using an average of many polls instead of relying on just one, but the fact remains that when you’re using fractions of a percentage point to decide who’s in and who’s out, the risk that someone who deserved to end up on the main stage will get shafted is real.

Here’s Seth Meyers begging Fox to include Perry in the main debate and give America the Trump/Perry barroom brawl we all deserve. Exit question via Byron York: To repeat a point I made yesterday, are we sure landing in the 5 p.m. debate isn’t a blessing in disguise? Not only do the participants not have to worry about being overshadowed by Trump, York argues, but the exchanges are more likely to be substantive and the smaller field will give each of them more chances to get noticed. Do well at the 5 p.m. debate and it’s possible your numbers will bounce high enough to land you on the main stage in the next one.