Something I’ve noticed about people’s responses to articles on this website is an apparent misunderstanding as to how my race ratings work. I haven’t really devoted a full article yet to how my rating system works, and as the election draws near, I want to rectify this.
For every election for the Electoral College, Senate, U.S. House of Representatives, and State Governor, as well as every presidential primary, since the beginning of 2019, I’ve assigned ratings using a 4-scale rating system. Each rating begins with one of the four words below, and ends with the party or candidate it applies to.
Currently Safe: The party or candidate indicated, if nothing changed between now and election day, would have a 100% chance of winning the election. I often abbreviate “Currently Safe” to “Safe” to reduce clutter on graphs and repetiton on articles.
Likely: The party or candidate indicated is a strong favorite to win the election, and it would be a surprise if that party or candidate lost. However, I can’t say with absolute, utter certainty that the election will go a certain way, so I leave the door open for an upset.
Lean: This is very much a contested election, and both (or in rare cases, all) parties/candidates have at least a decent chance to win. However, the party or candidate indicated has a definite leg up over all the others.
Toss-Up: No party or candidate has a clear advantage over the others. This race could go either way. These are the most exciting elections, and are the the namesake of this website. I make sure that no races are in Toss-Up by election day, because I don’t want to leave people guessing who really has the advantage on the big election day.
The reason I use this 4-scale system is that it’s the most accessible and understandable. It simplifies complicated electoral situations into neat, little phrases that are easy to digest. However, for those who seek to fully understand the rating system, one must recognize that there are a range of situations that fall into each category. For example, let’s look at the “Safe” rating. It’s important to note that a wide range of outcomes could come of Currently Safe races.
It may very well be that a race I rate like this is ultimately decided by less than 5 points. This never meant that the party indicated was in danger of losing the election, because they didn’t, there was just the possibility of the race being close. In some Safe races, it may be that the candidate that’s hopeless to win actually does better than Trump or Biden (depending on his or her party) in his or her region. A good example of where this could happen is South Carolina, where I currently have Lindsey Graham as a safe bet for re-election. This is despite Jaime Harrison raising towers of money from Democratic donors and ordinary voters alike. Jaime Harrison is one of the best congressional candidates in South Carolina in recent memory. But he’s running in the wrong state. South Carolina is very polarized along partisan and racial lines, and an establishment-leaning liberal like him won’t have robust enough support among Republicans to win in South Carolina, a Trump+14 state. Still, he may do better than Joe Biden in the state.
Other “Safe” races may be more likely to be 50 point blowouts, while in others still, there isn’t even formal opposition on the ballot to the incumbent party. A very wide range of electoral situations exist within the “Safe” category.
This doesn’t go away as we go down to the “Lean” and “Likely” categories. In fact, it is more important since either party has a chance of winning these races.
Some “Likely” races only exist as such because there’s a small chance that if everything breaks right for the other party, they might just win. A good example of this is NC-8. The Democratic candidate, Patricia Timmons-Goodson, is a respected, no-nonsense judge that was the first African-American on the North Carolina Supreme Court, and is on the U.S. Commission on Civil Rights. Many thought she’d be a good candidate to take on Richard Hudson, a generic Republican incumbent, in the Trump+9 district. However, she has struggled to raise sufficient funds. She isn’t particularly charismatic, nor does she have many detailed political positions. If she doesn’t get her act together by the next fundraising quarter, there’s no reason to expect her to have a chance of winning in this rural, southern, Trump-voting district, so this race would become Safe Republican. But I think she has potential as a candidate if she can get it together. That slight chance of this race being winnable for her keeps this race in Likely Republican for now.
Then you have the “Likely” races where an upset would be a surprise, but it’s not like the incumbent will ever be completely out of the woods either. A good example of this is NY-19. The incumbent Democrat is Antonio Delgado, a charismatic, young, African-American congressman. Some may consider him too liberal for the Trump+7 district, but he is very good at retail politics, at selling his ideas to voters while casting his Republican opponents as being outside the mainstream. The Republican candidate will most likely be Ola Hawatmeh, a Jordan-American fashion designer and entrepreneur. She’s a first time candidate with little political experience, and her fundraising pales in comparison to that of Delgado’s. Delgado should be fine, but you can never be too careful as a progressive Democrat in a red-trending district. He can’t take this election for granted, and he needs to make sure he still has the trust of the district’s many swing voters. This election has the potential to be decided within 5 points.
Finally, the Lean races. There seem to be 3 tiers of Lean races. Let’s look at three “Lean Biden” states and their differences.
Colorado is a Clinton +5 state, with a lot of rural and exurban working-class voters that have a history of voting Democratic. Curiously, in 2016, areas with a lot of these voters didn’t trend right for Trump that much compared to the Midwest. A possible explanation for this is the state’s higher level of bachelor’s degree attainment and greater diversity than the Rust Belt states. With America as polarized as it is, a vast majority of Americans will vote for the same party they voted for in 2016. So Biden has a good amount of insulation here. But if Trump can make up his 2016 deficit to win the state, it will be because the state’s working-class made a more pronounced shift to him, and he improved among rural Hispanics. However, as people become angrier at Trump for his COVID-19 response, Joe Biden could begin to lock this state up if he doesn’t trip up on an unforced gaffe.
Maine, a Clinton+3 state, is a bit more dicey for Biden. The state’s large working-class population raced to the right in 2016. I doubt Biden will have much more appeal than Clinton in small town America, and Maine seems to have an anti-big federal government attitude, so it remains very possible that Trump can capture this state. However, there are pockets of Trump voters here that could be convinced to vote for Demo0crats, as we saw in the 2018 elections. Also, Maine also has one of the highest median ages of any state, which should help Biden. Clinton won this state, so Biden starts off with an advantage, but I doubt his advantage will ever become insurmountable, like it could become in Colorado, barring a Trump implosion.
And then you have Michigan, a Trump+0.3 state that maybe should be in Toss-Up. It’s Lean Biden because I see an very slight edge for Biden. If I left this state as just purely a Toss-Up, I’d be lying about how I really felt about things. Joe Biden has demonstrated strength with the state’s large black population, which alone could put him over the top. Then you have polls that consistently show Biden leading Trump in the state. We know not the direction that the state’s white working class or the supposedly blue trending Oakland, Washtenaw, and Kent counties will trend, but I generally believe more things have to go wrong for Joe Biden to lose the state than Donald Trump.
There is no variation like this within the Toss-Up category. The only races I have as Toss-Up are the races where I genuinely can’t find a detectable, clear-cut advantage for one specific candidate over the other at the present time. It’s 50-50.
I hope that this guide helps you understand not only the differences between my ratings, but within my ratings as well. To clear up any further confusion, I decided take every race I’ve assigned a “Lean” rating to and sort them into one of the three tiers I mentioned above.
Hello There. I found your blog using msn. This is an extremely well written article.
I will make sure to bookmark it and return to read more of
your useful information. Thanks for the post. I’ll definitely comeback.
Thank you for your support.