Corrected 2:35 p.m. | Judging a candidate’s strength or weakness is one of the staples of political analysis, but too much of the speculation is done based on anecdotes and reputation.

The Vote Above Replacement, or VAR, metric quantifies how valuable (or how much of a liability) a particular candidate was to his or her party. It also helps us dig deeper into the Senate races that decided the 2020 elections and look ahead to some of the key contests next year.

VAR measures the strength of a political candidate relative to a typical candidate from their party within the same state. That initial benchmark is derived using Inside Elections’ Baseline metric, which captures each party’s political performance in a state or congressional district by combining all federal and state election results over the past four election cycles into a trimmed mean.

To calculate VAR, take any statewide candidate’s result and compare it to their party’s Baseline in that state for that cycle. It can even be done with the presidential race, by comparing each candidate’s performance within each state. An analysis of 2020 showed President Donald Trump underperformed an average GOP candidate in key battlegrounds. 

The 2020 flips

Democrats took control of the Senate thanks to four strong candidates. 

Retired astronaut Mark Kelly posted a +3.6 VAR in Arizona in defeating appointed GOP incumbent Martha McSally. With a -2.3 VAR, McSally actually improved on her 2018 performance, when she lost the state’s other Senate seat to Democrat Kyrsten Sinema and had a -4 VAR. 

In Colorado, former Gov. John Hickenlooper earned a +3.6 VAR against Republican Sen. Cory Gardner. Coming into the 2020 election cycle, GOP strategists were confident in Gardner’s ability to overperform the top of the ticket enough to win his second statewide race. Not only did Gardner post a modest +0.2 VAR in his initial 2014 race against Democratic Sen. Mark Udall, he slipped to a -1.7 VAR in 2020.

And finally in Georgia, Democrats Raphael Warnock (+5 VAR) and Jon Ossoff (+4.6) had solid performances in the Jan. 5 runoffs that secured control of the Senate for their party. In comparison, GOP Sens. Kelly Loeffler (-4.1 VAR) and David Perdue (-3.7) underperformed the average Republican candidate.

Some folks point to Joe Biden’s victory over Trump in Georgia in November as a way to undersell Warnock’s and Ossoff’s value as candidates. But that fails to consider eight years of Republicans winning statewide races in Georgia, which is accounted for in the Baseline metric. 

The lone Republican takeover was in Alabama, where Democratic Sen. Doug Jones had a +1.2 VAR. That was far less than his +10.5 VAR in the 2017 special election, when Jones got to run with a more favorable electorate and against embattled GOP nominee Roy Moore. In 2020, Republican Tommy Tuberville earned a mediocre -1.1 VAR in the race, but that was more than enough in a state with a GOP Baseline of 62.2 percent.  

2020 standouts, and duds

One of the top VAR’s in 2020 Senate races was posted by Maine Republican Susan Collins. Her +5.8 was the second-best among Republicans last cycle and just one piece of statistical evidence to support her impressive victory. She’s also the only Senate candidate in 69 races over the last two presidential cycles to split the ticket with the state’s presidential results. Her opponent Sara Gideon (-7.1 VAR) was one of the Democrats’ worst-performing Senate candidates in 2020.

Other Democratic challengers struggled or couldn’t outperform by enough to overcome the fundamental partisanship of their states. Cal Cunningham was ridiculed for his extramarital affair and grilling skills, leading to his -1.4 VAR in North Carolina. (GOP Sen. Tom Tillis had a -1.9 VAR but was saved by the state’s red tinge.) 

Democrat Jamie Harrison was backed by more than $100 million in campaign spending, but his +3.3 VAR wasn’t nearly enough to win in South Carolina. GOP Sen. Lindsey Graham actually had a worse VAR (-2.3), but he had the partisan lean of the state in his favor. 

Kentucky Democrat Amy McGrath raised a ridiculous amount of money as well and underperformed an average statewide candidate with a VAR of -3.3. Sen. Mitch McConnell’s team boasted about his nearly 20-point victory, but according to his VAR (+0.7), he did only slightly better than a typical candidate in Kentucky.

In Michigan, Republicans were convinced their nominee, John James, would overperform enough to win. He improved from his 2018 Senate loss, when he posted a -0.7 VAR, to +0.7, but that wasn’t enough to defeat Democratic Sen. Gary Peters (+0.6 VAR).

Republicans failed to win the open seat in New Mexico that Democrat Ben Ray Luján helped his party retain. But GOP weatherman Mark Ronchetti’s +2.5 VAR will do nothing to stop party strategists from trying to get him to run for something again. 

The worst-performing Senate candidate of the cycle was Nebraska Democrat Chris Janicek. The state party abandoned him and backed an alternative write-in candidate following a series of sexual misconduct scandals, contributing to his -13.1 VAR. With those circumstances and record turnout, GOP Sen. Ben Sasse received more votes in the general election than any candidate for any office in the state’s history, and helped him earn a 1.8 VAR. 

Looking ahead to 2022

The initial retirements from the Senate don’t appear to be an immediate problem based on past performance. GOP Sens. Richard Shelby of Alabama (+3.7) and Rob Portman of Ohio (+3.7) earned positive VARs in 2016, but Republicans have a significant advantage in each state and shouldn’t have trouble holding their seats. 

North Carolina is more competitive, but retiring Republican Richard M. Burr’s +0.4 VAR in 2016 doesn’t make him irreplaceable. The biggest GOP headache could be in Pennsylvania: Biden won the state’s electoral votes; Democrats have a 3.7-point advantage there, according to Baseline; and retiring Republican Sen. Patrick J. Toomey had a +2 VAR in 2016. 

Republicans are hoping a couple of their governors make Senate runs, but Montana Democrat Steve Bullock’s 2020 Senate race should be a cautionary tale about making the jump to federal office. The governor earned a +6.1 VAR in his 2016 reelection as governor but just a +2.1 VAR in his loss last year to GOP Sen. Steve Daines. Term-limited Maryland Republican Gov. Larry Hogan would likely need to pull off close to a +14 VAR in 2022 to unseat Democratic Sen. Chris Van Hollen. 

New Hampshire GOP Gov. Chris Sununu wouldn’t need to do quite as well against Democratic Sen. Maggie Hassan since the state is more competitive, with a post-2020 Democratic Baseline advantage of 49.9 percent to 47.1 percent. But Sununu should be under no illusions that it will be easy to replicate his +18.1 VAR from last year against an incumbent senator.

Nathan L. Gonzales is an elections analyst for CQ Roll Call.

Sen. Ben Sasse‘s VAR score has been corrected in this report.