**10%**Democrats gain control of the Senate in**10%**of our simulations.**22%**The Senate splits 50-50 in**22%**of our simulations. The vice president decides the balance of power in that case.**68%**Republicans keep control of the Senate in**68%**of our simulations.

## Possible Seat Counts

We simulated a Nov. 8 election 100 million times using our state-by-state probabilities. In **10.1 million** simulations, Democrats ended up with at least 51 seats. Therefore, we say Democrats have a **10.1 percent** chance of gaining control of the Senate.

Both independent senators, Bernie Sanders and Angus King, caucus with the Democratic Party. We count them as Democrats in our calculations.

## State-By-State Probabilities

The 2016 Senate consists of 54 Republicans and 46 Democrats. (The two independent senators caucus with Democrats.) Voters usually re-elect their incumbent senators, but some seats could flip to the other party.

If four Republican Senate seats flip to Democrats and there are no other changes, the 2017 Senate will be split 50-50.

### States Likely To Flip

Our algorithm suggests these states have at least a 50 percent chance of flipping.

Click or tap a race to see our calculations

##### Likely Winner

### States Less Likely To Flip

Our algorithm suggests these states have less than a 50 percent chance of flipping.

##### Likely Winner

## Methodology

We calculate the probable outcome for each individual Senate race, and then we use those probabilities to determine the likely seat counts on election night.

### 1. State-By-State Probabilities

We estimate the probability of a win in each Senate race using publicly available polls in the HuffPost Pollster database. We use Pollster’s Bayesian Kalman filter model to simulate 100,000 populations whose voting intentions correspond to the poll results. (We sample 5,000 of those simulations in our calculations, for speed.)

When there were fewer than five available polls in 2016 or fewer than two polls since July 2016, we used Cook Political Report ratings to estimate where the race stands.

We run the simulations out to Election Day, Nov. 8. Since we don’t have polling data for the future, the model assumes voter intentions generally continue along their current trajectories. The lack of poll data means that the outcomes of the races get less certain as time goes on, which moves the probability closer to 50 percent.

The model also estimates what proportion of voters is undecided according to the polls as of today. We penalize the leader’s win probability based on this formula: *percent undecided ÷ (leader’s poll average − runner-up’s poll average) × 100 percent*. The smaller the margin between candidates, the greater the chance that undecided voters will affect the outcome. We subtract that number from the overall probability of a candidate winning.

Download our state-by-state probabilities TSV file.

### 2. Likely Seat Counts

Finally, we simulated a Nov. 8 election 100 million times using the state-by-state probabilities. The proportion of times Democrats ended up with at least 51 seats is the probability of the Democrats gaining control of the Senate. The probability of a tie is the proportion of times the seat count landed at 50-50.

Download our likely seat counts TSV file.

Find out more about our methodology.