A 115% election is not a constitutional crisis—it is a clue about how odds are built.
Suppose an election board gives five candidates implied chances of 32%, 28%, 24%, 17%, and 14%. The candidates are mutually exclusive, yet the figures total 115%. At first glance, either probability has broken or someone has padded the numbers.
$2,750
200% Crypto Bonus PLUS 10% Gamblers Insurance with a minimum $100 deposit and a maximum of $1,500. Use Code: JOIN200 in the Cashier.
$250 + 100 Free Spins
All new players receive $250 in Free Bets and 100 Casino Spins. Join today and enter the promo code WELCOME in the cashier when you deposit $50 or more.
Get a 50% deposit bonus up to $1,000 with the code: CRYPTO1K. Valid on crypto deposits.
The catch is that displayed odds are usually prices, not pieces of one perfectly balanced forecast. Each price may include a bookmaker’s margin, reflect uneven betting demand, or come from contracts with different trading conditions. Converting every quote into an implied probability exposes those adjustments, so the total can rise well above 100%. The excess is therefore the useful mystery: how much comes from margin, and how much from the market itself?
- A total above 100% is commonly called the overround.
How odds become implied probabilities
- Decimal odds
Divide 1 by the quoted odds, then multiply by 100. Decimal odds of 2.50 imply 40%: 1 ÷ 2.50 = 0.40.
- Fractional odds
For odds of a/b, divide b by a + b. A 3/2 price implies 40%: 2 ÷ (3 + 2) = 0.40.
- Positive American odds
For +A, calculate 100 ÷ (A + 100). Odds of +150 therefore imply 40%.
- Negative American odds
For −A, use its absolute value and calculate A ÷ (A + 100). Odds of −150 imply 60%.
- Implied probability
This is the percentage encoded by the price—not necessarily the bookmaker’s pure estimate of the outcome. The same conversion also applies when reading novelty betting markets.
Implied probability should not be mistaken for a neutral, exact forecast. A quoted price may include the operator’s margin and adjustments for liabilities, expected betting demand, market uncertainty, or competitor pricing.
That is why converting every candidate’s odds and adding the results can produce a total above 100%. The excess is often called the overround, although its size does not reveal precisely how the margin is distributed across candidates.
The margin built into every price
Suppose two candidates are judged equally likely. Fair decimal odds would be 2.00 for each, implying 50% plus 50%. If both prices are shortened to 1.90, each implies 52.63%, so the market adds up to 105.26%.
The extra 5.26 percentage points are the overround, also called the vigorish, vig, or theoretical margin. Every outcome has effectively been made less generous than its estimated fair price. In this balanced example, equal stakes on both candidates would leave the operator with 5% of turnover: half the money would receive a 1.90 payout, while the other half would lose.
Terminology can be slippery. Overround describes the sum of implied probabilities above 100%; the corresponding theoretical hold is often calculated as overround ÷ total implied probability. Here, 5.26% divided by 105.26% produces a 5% theoretical hold. The distinction helps when examining how sportsbook hold works in practice.
Actual profit can be very different. Election betting rarely arrives in perfectly balanced amounts, and operators may carry a large liability on one candidate. Promotions, boosted odds, free bets, trading costs, and late price changes also affect the final figure. Most importantly, the result determines which liabilities are paid. Overround is a property of the posted prices; profit is an outcome of the bets actually taken.
Turning a 115% book into a 100% estimate
-
Start with four decimal prices
Suppose Candidates A, B, C, and D are priced at 2.00, 4.00, 5.00, and 5.00.
-
Convert each price
Using implied probability = 1 ÷ decimal odds, the calculations are: A = 1 ÷ 2.00 = 50%; B = 1 ÷ 4.00 = 25%; C and D = 1 ÷ 5.00 = 20% each.
-
Add the implied percentages
The total is 50% + 25% + 20% + 20% = 115%. The amount above 100% is a 15-percentage-point overround.
-
Normalize each candidate
Divide every implied percentage by the 115% total: A = 50 ÷ 115 = 43.48%; B = 25 ÷ 115 = 21.74%; C and D = 20 ÷ 115 = 17.39% each.
-
Check the adjusted total
The normalized figures add to 100%, apart from possible rounding. This removes the overround proportionally rather than treating the quoted odds as direct forecasts.
This estimate works only when the listed candidates form a complete, mutually exclusive field: exactly one can win, and every possible winner is included. Missing candidates, overlapping outcomes, or separate markets make the 100% rescaling misleading.
Why early prices carry extra padding
Pre-voting markets often remain open for months, leaving time for scandals, withdrawals, court rulings, health issues, and major polling shifts. That long settlement horizon exposes bookmakers to more uncertainty, so prices may be set cautiously and adjusted slowly.
The candidate field can also be unsettled. A prominent figure may not have declared, while listed candidates may later suspend their campaigns. Minor candidates are especially difficult to price because polling is sparse, name recognition can change quickly, and even a tiny estimated chance is hard to distinguish from virtually none.
Liquidity matters too. When little money has been wagered, bookmakers and exchanges receive less information from trading activity. Wider margins help protect against a few informed bets creating disproportionate liabilities.
The resulting total—perhaps 110%, 120%, or more—does not represent:
- unallocated votes;
- expected turnout;
- undecided voters; or
- people who will abstain.
Those concepts can affect forecasts, but they are not the mathematical source of overround. The excess comes from the prices attached to the listed outcomes. If “other candidate” or “no winner” is a valid possibility, it must appear as an explicit market outcome rather than being inferred from the amount above 100%.
Add only like-for-like outcomes
A percentage total is meaningful only when every price belongs to one market and exactly one listed outcome can win. Combining a candidate’s nomination odds with another candidate’s presidential odds produces a meaningless figure: those contracts answer different questions.
| Market | What usually determines settlement |
|---|---|
| Party nomination | The party’s official nominee |
| Presidency | The winner under the contract’s stated legal or certification standard |
| Popular vote | The highest vote total nationwide, or in a named jurisdiction |
| Electoral College | Electoral votes under the specified counting rules |
These outcomes can diverge. A candidate may win the popular vote but lose the presidency, while a nomination contract may settle months before either event.
Completeness also depends on the fine print. A candidate-only list may omit “other” or “the field,” leaving real winning possibilities outside the total. Some markets provide separate outcomes for a tie, no majority, or an unresolved result.
Replacements matter too. If a nominee withdraws, a contract naming that person may lose even when the replacement from the same party wins. Jurisdiction-specific definitions can also change what counts: certified totals, recounts, runoff rules, faithless electors, or deadlines for determining a winner.
Before adding percentages, the settlement rules should confirm that the options are both mutually exclusive and collectively exhaustive.
The rulebook decides the winner
An operator settles a market under its written rules, not necessarily when a network projects a winner, a candidate concedes, or an initial count shows a lead. Rules may instead name a certification authority, electoral vote, inauguration, or another official source. That distinction can keep funds locked long after the public considers the race finished.
When the ordinary result breaks down
Withdrawals, replacements, deaths, recounts, lawsuits, and delayed certification expose differences hidden by similar market titles. A bet on a named candidate may lose after that candidate withdraws, while a bet on the party may remain active. Another operator might void the candidate market or transfer it to a replacement—but only if its terms expressly allow that treatment.
Before combining prices, compare:
- the exact winning condition and jurisdiction;
- the named settlement source and deadline;
- whether concessions or projections count;
- rules for replacements, ties, disputes, and voids.
The settlement and withdrawal guides below cover these exceptional cases in more detail.
Four assumptions that distort the total
A quick audit for election odds
-
Capture prices at one time
Odds move quickly, especially after debates, withdrawals, and endorsements. Record each quote and timestamp before adding anything.
-
Read every contract
Confirm the office, election stage, closing date, settlement source, and treatment of replacements or withdrawals.
-
Build a complete outcome set
Include “other,” field, or no-candidate options where offered. Check that outcomes are both mutually exclusive and collectively exhaustive.
-
Convert prices consistently
Turn every decimal, fractional, or American quote into an implied percentage using the appropriate formula.
-
Add, then compare
The total measures overround only when the contracts match. Compare providers using quotes captured at similar times, and account separately for exchange commission.
-
Investigate unusually high totals
Check for stale prices, omitted outsiders, different deadlines, and incompatible settlement rules before treating the excess as margin.
A total above 100% usually reflects market cost and structure, not a forecasting contradiction. A very large excess deserves closer inspection: first verify timing, completeness, and contract terms; only then judge whether the margin itself is unusually wide.
