Are Democrats surging just in time to save their prospects for the mid-terms? That’s the narrative currently being pushed by the election analyst bros on social media and the press. A series of recent generic ballot polls have shown Democrats gaining on the GOP, and for the first time since last November, leftwing poll aggregator 538 shows Joe Biden’s party ahead.

There it is: for the first time since November, Democrats take the lead on the @FiveThirtyEight polling average for the generic ballot.

Environment on the day of the Dobbs decision: R+2.3
Environment 6 weeks later: D+0.1 pic.twitter.com/m7yc8RK033

— Lakshya Jain (@lxeagle17) August 5, 2022

But while election bro Twitter rejoices, I’d caution them that disappointment lurks around the corner. If you check the calendar, it is currently August. Do you know who led the generic ballot on August 6th of 2014? Democrats led it by almost 2 points, far more than their current lead on the same day in 2022. What was the final result? It was Republicans by 5.7 points. In other words, the polling difference between August 6th of 2014 and the real results that November was over 7 points. And Democrats are celebrating an August generic ballot average lead of 0.1 percent.

If you look at 2016 and 2020, you will find the same types of discrepancies between the generic ballot and the actual results. Why? Because generic ballot polling historically underestimates Republican support. Further, Democrats typically do better in the summer than in the fall. By October of 2014, the GOP had taken the generic ballot lead in the aggregate.

All that to say that it’s way too early for anyone to be suggesting that Democrats have saved themselves. The fundamentals are still terrible, and the first mid-term for a new president is historically awful for the incumbent party. If Republicans can best the average by several points in 2014, there’s every reason to believe a Democrat lead of 0.1 percent in 2022 is meaningless.

Then there’s the Senate polling. Most of the primaries have just ended, and very little actual campaigning has started. Yet, a series of partisan polls have Democrats believing they are going to sweep the Senate toss-ups, including in Ohio.

Polling in Ohio has been pretty questionable in quality, but here’s one from a decent firm (Impact Research) instead of SurveyMonkey:

Tim Ryan: 48
JD Vance: 45

July 21-28, 800 LV, MOE +/- 3.5.

Source: https://t.co/IvZeOSDwo3

— Lakshya Jain (@lxeagle17) August 6, 2022

Median, last 3 Senate polls:
PA D+11%
OH D+6%
NV D+6%
AZ D+5% (50th for D's/51st for R's)
GA D+3%
NH D+3-4%*
WI D+2%*
NC R+3%
MO R+6%*
FL R+6%

So for R's to reach threshold to take the Senate, conditions would have to move 5 points in their direction. (e.g. "meta-margin" = 5%)

— Sam Wang (@SamWangPhD) August 5, 2022

While the generic ballot has been questionable in the past, Senate polling has been far worse. For example, the aggregates in 2018 missed the Florida senate race by 2.6 percent. They missed the Indiana race by 6.6 percent. Missouri’s race was off by 5.4 percent. In fact, in all but a single toss-up (Nevada), the polling systematically underestimated Republican support. The same thing happened again in 2020,  but worse. Every single toss-up race ended up having a polling aggregate that underestimated actual GOP support.

And to make my point even starker, there has been precious little legitimate polling done as of this writing. For example, there have been zero non-partisan polls done in Ohio. Nearly all the polls used above by Sam Wang to make his aggregates were done by scam PACs like Center Street and other hired guns putting out numbers to push fundraising. There’s absolutely nothing of value to be gained by his post.

Here’s the reality. The Democrats are getting a summer polling bump, something they’ve often seen in past cycles because most people aren’t paying attention right now. Mid-term campaigns don’t even really start until September. That’s when the big ad buys start, that’s when the debates begin, and that’s when rallies and other events become predominant. If Republicans are still trailing by October, maybe there’s some cause for concern, though past polling errors say even then, not much. But anyone taking a 0.1 percent generic ballot for Democrats in August as a sign they are going to come out triumphant is just showing how gullible they are.