Is An RFK Jr. Effect Now Emerging?

Four new polls suggest that something may be taking shape

On April 13 I wrote a piece called “Is An RFK Jr. Effect Actually Happening?”. In this article I examined and analyzed a month’s worth of general election presidential polling of Trump, Biden, and RFK Jr. My goal was to determine if there was an “RFK Jr. effect”. I wanted to see if it was possible to discern any tendencies of him pulling respondents away from Trump or Biden, or having no net effect on either.

My conclusion was not too hard to come by: [spoiler alert] There was no discernable RFK Jr. effect. He sometimes seemed to have an effect on Trump, sometimes Biden, all to varying and random degrees. And many of the polls that showed Biden losing ground on Trump when you threw in RFK Jr. were really showing an effect from the addition of West and Stein into the mix. There were not nearly as many polls that included RFK Jr. that did not also include West and Stein, but the ones that did only include RFK Jr. still had no obvious net effect.

I’m glad I wrote that article then, because had I waited for a week or so, I may have had to scrap it all and readjust to a new reality. Four new polls just came out, and they all suggest that there may be an RFK Jr. effect emerging. The thing that makes these new polls so compelling is that, although their overall head-to-head results are quite different—the RFK Jr. effect is apparent and similar in all of them.

So it’s hard to just ignore them or dismiss them as polling noise. It’s still a bit too early to say there’s a definite trend, but four completely different polls within the same week, showing different overall results, but very similar RFK Jr effects, should be noted. Let’s examine the results.

Here’s a table I made to breakdown the 4 recent polls I’m referring to:

At first, it might look a little complicated, but it’s really pretty simple. The 2 left-most columns are the overall total results for Trump and Biden, head-to-head (H2H) and when 3rd party candidates are included (w/3rdP). The 3rd Party candidates are always RFK Jr., Cornel West, and Jill Stein in these cases.

The next column is the total of only the 3rd party candidates, which is broken out between RFK Jr. (RFK) and the combination of West’s and Stein’s totals (W-S).

The next four columns are the total point drop from head-to-head results to when 3rd Party is included for each of Biden and Trump (B or T Total Drop), and then the drops for each due solely to RFK Jr. (B or T RFK Drop).

Then the last column is difference between their drop totals due solely to RFK Jr. (RFK Drop Difference), which is essentially the RFK Jr. effect. It’s highlighted in red to show that Trump is on the wrong side of this effect. Since the difference in drops due solely to RFK Jr. is significantly larger for Trump in each case, and by several points at that. In other words, RFK Jr. appears to be pulling support more away from Trump than Biden.

This is apparent in two of the polls, as Biden either extends his lead when the 3rd party candidates are included (Marist) or takes over the lead (NBC). In the other two, it may not be as apparent, because in one Trump’s lead is extended (CNN), and in the other they remain tied (Quinnipiac). But Biden’s point drop in these cases is largely due to West and Stein, not RFK Jr.

Now, here is where I make a crude—although, I believe, safe—assumption. It’s the same assumption I made in my previous article on this subject. I assigned all of West’s and Stein’s support to Biden. So I took the total points of West and Stein and assume that their total is all coming from Biden’s total.

For example, in the Marist poll Biden had an 8-point drop when the 3rd party candidates were included. Since West and Stein’s total was 4 points, I simply subtracted 4 from 8, and the result of 4 is the point drop due solely to RFK Jr.

For Trump, you may have noticed that his Total Drop and RFK Drop are identical. This is because I’m assuming that none of his support is going to West or Stein, and that all of his drop is accounted for by RFK Jr.

This probably isn’t 100% accurate; like I said, it’s a crude assumption. However, it is not feasible to me that a meaningful amount of West and Stein support is coming from Trump. This doesn’t mean zero amount, but just not a meaningful amount. Therefore, it can be completely ignored for these purposes. If you want to argue for why there should be some inclusion of West and Stein support for Trump, feel free to comment and try to convince me. But even if we just throw in a point for Trump to account for this, you still have the RFK Jr effect, it’s just a bit diminished.

In my prior article, I didn’t think the RFK Drop Difference in those cases was compelling enough to label it an “RFK Jr. effect”. But I could have argued that it was. It’s just that most polls have a margin of error in the 2-3 point range, so these differences didn’t interest me.

I didn’t put the data from my previous analysis into a chart at the time I wrote it, but here it is now in the same format as above, for comparison’s sake:

As you can see, the RFK effect, as measured in the “RFK Drop Difference” column, is significantly diminished in these earlier polls. At the time I could have technically argued that the effect seems to be more against Trump. But in the context of all the polls I analyzed then, plus assuming a margin of error of a couple points, this did not seem significant at all, especially with there being a tie in there. I stand by my conclusions in that article.

However, in the context of the recent polling, it seems that a trend may possibly have been emerging at that time, it just wasn’t obvious then, but is more obvious now with the updated data.

The usual caveats apply here. It’s still very early to draw any major conclusions from polling. It’s best to mostly ignore the polls at this stage in the election cycle. Plus, there may be better, more accurate internal polling being done by the campaigns that show something completely different. This is just based on the current publicly available data.

But as far as media reports and punditry goes, this is the data we have, and if anyone is going to draw a conclusion about the effect RFK Jr. is having based on current polling, they’d have to say it looks like there is an RFK Jr. effect and RFK Jr. is hurting Trump more than Biden.

Respected Election Expert Joe Trippi Disagrees, but I have questions

It is worth mentioning that not everyone agrees with this assessment, regardless of its evidence in the polling. Election expert Joe Trippi has an interesting theory about an RFK Jr. effect. I’ll try my best to summarize here.

On his podcast That Trippi Show, plus other podcast interviews he has done¹, he explains that based on his focus groups and previous election results analysis, there is a fixed ceiling for Trump, which is about 45-46%. He argues that that really never changes, and that anyone that truly supports Trump will not be pried away from him. Therefore, it can be expected that 54-55% are left over and available to Biden in 2024.

But he will be vying for these voters along with any other candidates, such as RFK Jr., West, and Stein. Therefore, he equates RFK Jr. with only possibly damaging Biden’s prospects, not Trump’s. According to him, Trump is essentially 46%, and Biden will be anywhere from 46-54% depending on how well he can garner the non-Trump support.

I can understand this argument and am definitely not inclined to counter Trippi’s take on the election, as he has decades’ worth of knowledge and experience.

But I do have a couple issues with his theory that make me wonder about its accuracy.

First, in these same podcasts he often brings up the fact that Nikki Haley voters have shown themselves to be significant players in the Republican primaries, and that they therefore could doom Trump in November, as many of them have indicated they will not vote for Trump. Now Trippi does lump these people into the 54-55% that is up for grabs for Biden. But I’m not sure that’s completely accurate.

Wouldn’t it be assumed that many of these Haley/non-Trump voters actually voted for Trump in 2020 and 2016, as they were largely Republican voters? If so, this would mean that Trump’s ceiling would be even less than 46%, since that’s essentially the total he got in each of the last two elections. Keep in mind that the 2020 voters didn’t have Jan 6 or Trump’s criminal legal issues to think about when casting their votes.

It seems likely that more people will be flocking away from him, or at least be inclined to do so. In that case, they may prefer an option like RFK Jr. for no other reason than they won’t be voting for Biden. RFK Jr. does have some very MAGA-ish opinions about some issues, after all.

In addition, we are 8 years away from 2016, and 4 years removed from 2020. The electorate would be assumed to have shifted to some degree, even if it’s very minimal. It’s feasible that given the history with Trump and the experience that the country has with him now has shaken some of his support. He’ll always have some amount of die-hard unshakable support, but it’s reasonable to assume that his support is getting worn on the edges.

It wouldn’t surprise me if Trippi is right on the money with this, but also wouldn’t shock me if he’s off the mark. I think a scenario where, as we get closer to the election the polls show the head-to-head numbers solidifying around the binary choice being between Trump and Biden, and RFK Jr. therefore becoming less and less significant, is very possible. And this would probably result in Trippi being accurate in his assessment.

The CNN poll is an outlier

The other notable thing is that the CNN poll I cited here is an outlier and could probably be thrown away and ignored to most practical purposes. A 6-point lead for Trump is not realistic, nor is his lead among women, younger voters, and his reduced margins with non-white voters in this poll. Sometimes outliers happen and you can’t take them seriously.

I did think it was worth including in this analysis, however, because the RFK Jr. effect is consistent here with the other polls. I thought this might help indicate an emerging trend, even if the poll is way off on other factors. I could have ignored it and just included the three others, and still would have confidently come to the same conclusion.

Overall RFK Jr. Support Update

Another thing I looked at in my previous RFK Jr. effect article was his overall support and how it had been trending, along with overall 3rd party support when including West and Stein. Using data from monthly Quinnipiac polls from Oct 2023 through March 2024, it could be seen in a simple bar graph that support for RFK Jr. was steadily declining. But, interestingly, while overall 3rd party support had some decline, it had somewhat stabilized in the most recent few months, suggesting that an overall appetite for 3rd party candidates was fairly stable, even as RFK Jr’s support was declining. In other words, support for West and Stein were increasing as support for RFK Jr. was decreasing.

Well, this data took a bit of a turn as of Quinnipiac’s latest poll in April, conducted 4/18-4/22. Here’s the same two graphs as last time, with April’s updated data added in:

As you can see, there’s a bump up in April for RFK Jr. and for 3rd party overall, the highest levels since December. Notably, West and Stein support decline a bit. It’s hard to make out what this means, if anything, for the same caveats as I mentioned above regarding polls in general.

But it’s notable that as of now, RFK Jr. and 3rd party options are remaining stable, and even increasing a tick. Perhaps Trump’s criminal trials have something to do with this, as maybe some portion of his voters have turned away from him and are looking to other options. In fact, this could be the primary driver of the RFK Jr. effect in the first place.

We won’t know for sure what effect the Trump prosecutions are having for another few months. And, of course, the poll data can change at any time in the future. But it can now be said, based on the publicly available data, that there is an RFK Jr. effect, and it’s pulling support away from Donald Trump.

1. For example, Trippi’s 4/24 interview with Greg Sargent on The New Republic podcast The Daily Blast