What this Election Swingometer does
This Election Swingometer estimates how a uniform change in vote share might alter the margin between a leading party and its runner-up in one constituency. It uses the supplied previous shares, an estimated current share for the leading party, and an applied swing factor. The output is a modeled margin in percentage points. A positive result means the modeled leading party remains ahead. A negative result means the runner-up would be ahead under the calculation.
The model uses a deliberately simple assumption. It treats the change in the leading party’s share as a swing that is matched by an equal and opposite change in the runner-up’s share. That means the gap between them moves twice as much as the one-party swing. The calculator describes a scenario, not a forecast of an election result. It does not account for turnout, third parties, changing boundaries, local campaigning, or differences among constituencies.
For example, suppose the leading party previously received 40% and the runner-up received 35%. If the leading party’s estimated share is now 38%, the change is minus 2 percentage points. At a 100% swing factor, the model gives the leader 38% and the runner-up 37%. The resulting margin is 1 percentage point. The arithmetic uses percentage points, which measure the difference between percentages.
How to use the Election Swingometer
Enter the previous vote shares for the two parties being compared. Use the same election and constituency for both values. Enter the leading party’s current estimated share as a percentage from 0 to 100. The fourth input controls what share of the change to apply. A value of 100 applies the full change. A value of 50 applies half of it. A value of zero holds both modeled shares at their previous values.
- Leading party previous vote share is the starting share for the party that led in the prior comparison.
- Leading party current estimated vote share sets the scenario’s updated share for that party.
- Runner-up previous vote share is the runner-up’s starting point.
- Applied share of national swing scales the change for a scenario. It is not independently measured by the calculator.
Check that every share uses the same basis. Rounded published figures can produce small differences, and the shares entered need not add to 100 because other parties may have received votes. Try alternative assumptions if you want to compare scenarios. Keep a note of the values used so readers can understand how you reached the displayed margin.
How the Election Swingometer formula works
The calculator first finds the change in the leading party’s share by subtracting its previous share from its current estimated share. It multiplies that change by the applied swing factor divided by 100. The adjusted change is added to the leader’s previous share and subtracted from the runner-up’s previous share. Finally, the runner-up’s modeled share is subtracted from the leader’s modeled share.
In compact form, the margin equals the leader’s previous share plus the adjusted change, minus the runner-up’s previous share after the same adjusted change has been deducted. If the adjusted change is minus 2 points, the leader loses 2 points and the runner-up gains 2 points. The margin therefore falls by 4 points. This is the defining assumption of the model, sometimes called a uniform two-party swing.
The output is a difference in percentage points, not a percentage change. A modeled margin of 1.00 pp means the leading party is one percentage point ahead of the runner-up in this two-party comparison. It does not indicate the probability of winning, the number of votes involved, or the size of an overall majority.
Use cases for the Election Swingometer
A swingometer can help make an election discussion more concrete. Civic groups, students, journalists, and local residents can use it to see how a stated assumption changes a constituency’s notional margin. It can support a classroom exercise comparing a uniform swing with actual results, or help readers understand why a narrow previous margin may be sensitive to a modest movement in support.
One useful approach is to hold the previous results constant and test several current-share estimates. Another is to keep the estimated change fixed and compare applied factors, such as 50% and 100%. These are scenario exercises. The factor should be explained as an assumption unless it comes from a clearly identified source and method.
When presenting the result, include the starting election, constituency, parties, estimated share, and swing factor. Show the assumptions alongside the number. That context helps prevent a modeled margin from being mistaken for a reported poll or an official projection.
Other factors to consider when calculating election swing
Real elections rarely produce the same movement everywhere. Constituencies can differ in demographics, candidate appeal, campaign activity, local issues, and turnout. Voters may move among several parties, abstain, or participate for the first time. A two-party calculation compresses these changes into one comparison and cannot explain where votes came from or where they went.
Vote shares also depend on the denominator. A party can gain share while receiving fewer votes if turnout changes, and raw vote totals cannot be inferred from share alone. Boundary changes make comparisons especially difficult because the population and electorate may differ. If a seat’s boundaries changed, use a properly adjusted historical result where available and clearly describe its source.
Polling estimates bring sampling uncertainty, fieldwork dates, weighting choices, and other limits. A national estimate does not automatically translate into an identical local change. Treat this calculator as an arithmetic aid for exploring a stated assumption. For credible analysis, compare the scenario with constituency-level history, multiple data sources, and the relevant election context.
Election Swingometer FAQ
What does a positive modeled margin mean?
A positive result means the party entered as the leader still has the higher modeled share than the runner-up after applying the entered change. The result is the difference in percentage points between those two modeled shares.
Why does a small swing change the margin by twice as much?
The formula assumes the leader’s share falls by a given amount while the runner-up’s share rises by the same amount, or vice versa. Because the parties move in opposite directions, their gap changes by twice the one-party movement.
Can the calculator predict which party will win?
No. It calculates a margin under the values and uniform-swing assumption supplied. Actual results can differ because of turnout, local variation, other parties, candidates, and many other factors.
What does the applied swing factor represent?
It scales the estimated change. At 100%, the full change between the previous and current estimated leader share is applied. At 50%, half is applied. It is a scenario control and has no built-in evidential basis.
Do the vote shares have to total 100%?
No. The inputs cover only the leading party and runner-up. Other parties and independent candidates can account for the remaining share. Since the model only moves the two selected parties, its result should be read as a simplified comparison.