Account
Language
Time Zone
COEF
Half Time Predictions Yesterday
Review yesterday's Half Time football predictions on Betiball and compare the original Home (1), Draw (X) and Away (2) probabilities with the actual score when the first half ended. Each completed fixture provides a historical record of what the model expected before kick-off, which team was leading at the interval, or whether the match was level, and whether the leading Half Time prediction was correct.
Use the results below to review yesterday's first-half forecasts or explore all football predictions from yesterday, yesterday's 1X2 results, Half Time predictions today, Half Time predictions tomorrow and all Half Time predictions.
2026-09-18
Quy Nhon United vs Ho Chi Minh City - Prediction & Odds
2026-09-18
Brentford vs Chelsea - Prediction & Odds
2026-09-18
RCD Espanyol vs Elche - Prediction & Odds
2026-09-18
Bayern Munchen vs Union Berlin - Prediction & Odds
2026-09-18
Philippines U23 vs Vietnam U23 - Prediction & Odds
2026-09-18
Kuwait U23 vs Uzbekistan U23 - Prediction & Odds
2026-09-18
Saudi Arabia U23 vs Qatar U23 - Prediction & Odds
2026-09-18
Monza vs Sassuolo - Prediction & Odds
2026-09-18
Monaco vs Lens - Prediction & Odds
2026-09-18
Arema FC vs Persik Kediri - Prediction & Odds
2026-09-18
Madura United vs PSIM Yogyakarta - Prediction & Odds
2026-09-18
New York City FC vs New York Red Bulls - Prediction & Odds
2026-09-18
Ayutthaya United vs Uthai Thani FC - Prediction & Odds
2026-09-18
Yate Town vs Chippenham Town - Prediction & Odds
2026-09-18
Worcester City vs Stratford Town - Prediction & Odds
2026-09-18
Zhejiang Professional FC vs Wuhan Three Towns - Prediction & Odds
2026-09-18
Puebla vs CF Atlante - Prediction & Odds
2026-09-18
FC Juarez vs Tigres UANL - Prediction & Odds
2026-09-18
Bristol City vs Watford - Prediction & Odds
2026-09-18
Groningen vs PEC Zwolle - Prediction & Odds
Yesterday's Half Time Prediction Results
This page reviews completed first-half predictions rather than forecasting upcoming matches. For each fixture, the original Half Time Home, Draw and Away probabilities should remain visible beside the actual interval score and settled prediction result.
The Half Time market uses three possible outcomes:
| HT Result | Meaning | Example Half-Time Scores |
|---|---|---|
| 1 | Home team leads at half-time | 1-0, 2-0, 2-1 |
| X | Teams are level at half-time | 0-0, 1-1, 2-2 |
| 2 | Away team leads at half-time | 0-1, 0-2, 1-2 |
Predicted Half-Time Result vs Actual HT Score
A useful historical results page should retain the information that existed before kick-off and add the actual first-half outcome separately after the match.
| Field | Historical Meaning |
|---|---|
| HT Home % | Pre-match probability of the home team leading at half-time |
| HT Draw % | Pre-match probability of the teams being level at half-time |
| HT Away % | Pre-match probability of the away team leading at half-time |
| HT Prediction | The highest-ranked first-half outcome |
| Half-Time Score | The score when the opening period ended |
| Actual HT Result | 1, X or 2 |
| Prediction Result | Correct or Incorrect |
What Counts as a Correct Half Time Prediction?
The leading pre-match Half Time selection must match the actual result at the interval.
Correct Home prediction
If Home was the highest-ranked outcome and the home team reached the interval ahead, the prediction is correct.
Correct Draw prediction
If X was the leading selection and the teams were level when the first half ended, the prediction is correct. The score can be 0-0, 1-1 or another level score.
Correct Away prediction
If Away carried the highest probability and the visiting side was ahead at the interval, the selection is correct.
The Full-Time Result Does Not Change the Half-Time Settlement
One of the most important distinctions on this page is that the final match result cannot retrospectively alter the Half Time outcome.
| HT Score | FT Score | Half Time Result |
|---|---|---|
| 1-0 | 1-2 | 1 |
| 0-0 | 2-0 | X |
| 0-1 | 2-1 | 2 |
| 1-1 | 1-3 | X |
The second half belongs to a different prediction problem. For the final Home, Draw or Away result, review yesterday's 1X2 predictions.
Why All Three Original HT Probabilities Should Stay Visible
A simple Correct or Incorrect badge does not show how uncertain the forecast was before kick-off.
Consider two historical predictions where Home was selected:
| Original Forecast | HT Result | Interpretation |
|---|---|---|
| Home 61% / Draw 29% / Away 10% | X | A relatively strong Home forecast that missed |
| Home 37% / Draw 35% / Away 28% | X | A highly uncertain forecast where Draw was close behind |
Both leading selections are incorrect, but the original probability distributions tell very different stories.
The Probability of the Actual Half-Time Result Adds Context
For completed fixtures, Betiball can show the probability originally assigned to the outcome that eventually occurred.
| Pre-Match HT Forecast | HT Score | Actual HT Result | Original Probability |
|---|---|---|---|
| Home 55% / Draw 32% / Away 13% | 0-0 | X | 32% |
| Home 34% / Draw 42% / Away 24% | 1-0 | 1 | 34% |
These are illustrative examples rather than current Betiball results.
0-0 at Half-Time Deserves Its Own Historical Metric
A large proportion of first-half Draw outcomes can come from goalless opening periods, so a separate 0-0 HT metric can add information beyond the generic X result.
Predicted 0-0 probability
If available, this shows how much of the model's HT Draw expectation came specifically from a goalless first half.
Actual 0-0 rate
Historical results can show how often fixtures expected to be low-scoring in the opening period actually reached the interval without a goal.
1-1 and other HT draws
A correct Draw prediction does not necessarily indicate a quiet first half. A 1-1 interval is still X but represents a different scoring environment.
First-Half Goal Totals Help Explain the Result
Recording the number of goals scored before the interval provides additional context around the Home, Draw or Away result.
| HT Score | First-Half Goals | HT Result |
|---|---|---|
| 0-0 | 0 | X |
| 1-0 | 1 | 1 |
| 0-1 | 1 | 2 |
| 1-1 | 2 | X |
| 2-1 | 3 | 1 |
Expected First-Half Goals Can Be Audited Against Reality
If Betiball stores an expected first-half goal value before kick-off, yesterday's completed fixtures provide a natural audit dataset.
Expected HT goals
This describes the model's central estimate of first-half scoring rather than one guaranteed score.
Actual HT goals
The completed interval score provides the realised first-half total.
Expected vs actual
The difference can be analysed over large samples to see whether the model systematically expects too much or too little first-half scoring.
One individual match should not be used as evidence of systematic bias.
Half-Time Draw Accuracy Should Be Reported Separately
The Draw is a particularly important outcome in first-half forecasting because teams have less time to create a decisive score difference.
A single overall accuracy figure can hide whether the model is good at identifying level intervals.
| Prediction Type | Useful Historical Measures |
|---|---|
| HT Home | Picks, correct picks, accuracy and average probability |
| HT Draw | Picks, correct picks, accuracy and average probability |
| HT Away | Picks, correct picks, accuracy and average probability |
Prediction Accuracy Needs Sample Size
Yesterday's Half Time accuracy becomes more meaningful when users can see how many settled fixtures sit behind the percentage.
Total predictions
This should include every eligible Half Time prediction published under the same methodology.
Correct predictions
All successful Home, Draw and Away selections can contribute to the daily total.
Incorrect predictions
Unsuccessful forecasts should remain visible and form part of the calculation.
Unsettled fixtures
Matches without a valid Half Time result should remain outside the settled denominator.
Confidence Gap Shows the Strength of the Original Preference
The difference between the highest and second-highest Half Time probabilities can help users distinguish strong model preferences from near-balanced predictions.
A simple calculation is:
Confidence Gap = Highest HT Probability − Second-Highest HT Probability
| HT Distribution | Confidence Gap |
|---|---|
| Home 59% / Draw 30% / Away 11% | 29 percentage points |
| Home 36% / Draw 35% / Away 29% | 1 percentage point |
Both examples technically favour Home, but the strength of that preference is completely different.
Accuracy by Probability Band Is More Informative Than One Daily Number
Historical predictions can be grouped by the original probability of the leading HT selection.
| Leading Probability | Historical Use |
|---|---|
| 34% to 39% | Highly balanced three-way forecasts |
| 40% to 49% | Moderate leading preference |
| 50% to 59% | Clearer first-half preference |
| 60%+ | Strongest model preference |
The ranges are useful for reporting only. Their real value depends on whether observed accuracy increases as predicted probability rises.
Calibration Is a Better Test Than Highlighting Winning Picks
A probability model should be evaluated on whether its percentages correspond with observed outcomes over sufficiently large samples.
If HT Draw outcomes assigned around 50% occur approximately half the time across a suitable historical sample, that supports calibration. If they occur far less frequently, the model may be overestimating Draw probability.
The same analysis can be performed independently for Home and Away first-half outcomes.
One Day Cannot Establish Long-Term Half-Time Performance
Yesterday's fixtures provide transparency, but the sample can be affected by natural football variance.
An early penalty, dismissal, deflection or exceptional finish can change the Half Time result even when the original probability distribution was reasonable.
Daily performance therefore becomes more useful when accompanied by larger rolling samples.
7-day accuracy
This provides a short-term view across several matchdays.
30-day accuracy
A larger sample reduces the influence of individual unusual fixtures.
90-day accuracy
This can provide stronger evidence where the underlying model has remained sufficiently consistent.
First-Half Performance Can Differ by Competition
Leagues and tournaments can have different early-scoring environments, tactical styles and levels of competitive balance.
HT Draw baseline
Some competitions may produce level first halves more frequently than others.
Early home advantage
The frequency with which home teams establish first-half leads can vary across leagues.
Early away leads
Away-team performance should also be measured against the relevant competition rather than a single global baseline.
Where sample size permits, Betiball can therefore report historical Half Time results by league.
Home and Away First-Half Results Should Be Audited Independently
A model may be well calibrated for Home HT selections while overestimating or underestimating Away leads.
Home predicted vs Home actual
Track how often Home leads occurred at different predicted probability levels.
Away predicted vs Away actual
The same analysis can reveal whether Away first-half advantages are being modelled consistently.
Draw as the third outcome
Because every fixture has three possible interval results, calibration should consider the complete Home/Draw/Away probability distribution rather than only the leading pick.
Yesterday's First-Half Score Should Never Be Rewritten Using Full-Time Information
Historical transparency requires Betiball to preserve the forecast that existed before kick-off.
Keep the original probabilities
Do not recalculate yesterday's HT percentages after the match result is known.
Keep the original leading prediction
A losing selection should remain visible.
Add actual HT score separately
The result is settlement information, not a replacement for the prediction.
Preserve the timestamp
This records when the final pre-match forecast was produced.
Half Time and Half Time/Full Time Results Are Different
Half Time requires only the result at the interval. Half Time/Full Time requires both the interval result and the final outcome.
| Market | HT Result Required | FT Result Required |
|---|---|---|
| Half Time | Yes | No |
| Half Time/Full Time | Yes | Yes |
If a match is 0-0 at the interval and finishes 2-0, the Half Time result is X. The HT/FT result is X/1.
For the two-stage historical result, see Half Time/Full Time Predictions Yesterday.
Half Time and Live 1X2 Also Measure Different Events
A Half Time forecast predicts the result when the opening period ends. Live 1X2 estimates the eventual match result from an in-play state.
For example, a fixture reaching the interval at 0-0 has an actual Half Time result of X. At that moment, a Live 1X2 model could still favour Home to win the match.
Review historical in-play probabilities at Live 1X2 Predictions Yesterday.
First-Half Goals and Full-Match Under/Over Are Separate Outcomes
A goalless first half does not guarantee a low-scoring final result. Likewise, an open first period does not guarantee the match will continue at the same rate after the break.
A fixture can be 0-0 at half-time and finish 2-2. It can also reach the interval at 2-1 and end with no further goals.
For yesterday's full-match goal-total results, visit Under/Over 2.5 Predictions Yesterday.
Postponed, Abandoned and Suspended Matches Need Separate Treatment
A historical accuracy table should include only fixtures with a valid result for the prediction being evaluated.
Postponed before kick-off
No first half took place, so there is no Half Time result to settle.
Abandoned during the first half
The score at abandonment should not automatically be treated as the Half Time result.
Abandoned after half-time
The actual first-half score may exist, but the treatment of the prediction should follow Betiball's clearly stated historical-results methodology.
Suspended match
A fixture expected to resume should remain clearly distinguished from a normally completed match.
The Half-Time Result Includes First-Half Added Time
The interval result is determined when the referee ends the first period, not mechanically when the match clock first displays 45 minutes.
This matters because goals scored during first-half added time still belong to the opening period and can change a Home, Draw or Away Half Time result.
Betiball should therefore use the officially recorded interval score from its match-data source rather than infer the HT result from an arbitrary 45:00 snapshot.
Browse Half Time Predictions Across Matchdays
Review Other Prediction Results from Yesterday
| Prediction Market | Historical Result |
|---|---|
| 1X2 | Home, Draw and Away full-time results |
| Live 1X2 | Historical in-play probability changes |
| Under/Over 2.5 | Full-match total-goal results |
| Half Time/Full Time | HT-to-FT result combinations |
| BTTS | Both Teams to Score results |
| Double Chance | 1X, X2 and 12 prediction results |
| Asian Handicap | Handicap settlement results |
| Goalscorers | Player scoring prediction results |
| Corners | Corner prediction results |
| Cards | Card prediction results |
How Yesterday's Half Time Accuracy Should Be Reported
A transparent record should use the same settlement and accuracy methodology across dates, competitions and prediction outcomes.
Include every eligible settled prediction
Do not calculate headline accuracy from a selected subset of successful fixtures.
Preserve all three probabilities
Home, Draw and Away should remain visible even if only one was the leading prediction.
Preserve prediction revisions
If team news caused the probabilities to change before kick-off, the final pre-match version can be used for headline settlement while earlier versions remain available for auditing.
Keep losing predictions visible
Incorrect selections are necessary for users to understand genuine historical model performance.
Show sample size with every accuracy figure
The number of settled predictions should accompany overall and outcome-specific accuracy rates.
Common Questions About Yesterday's Half Time Predictions
What does a Half Time prediction mean?
It predicts whether the home team will lead, the teams will be level or the away team will lead when the first period ends.
What does 1 mean at Half Time?
The home team is ahead when the first half ends.
What does X mean at Half Time?
The teams are level at the interval.
What does 2 mean at Half Time?
The away team is ahead at the interval.
Does 0-0 count as a Half Time Draw?
Yes. The result is X because the teams are level.
Does 1-1 also count as X?
Yes. Any level interval score is a Draw.
Does a second-half comeback change the Half Time result?
No. Once the opening period has ended, the Half Time outcome is fixed.
Can a 60% HT prediction still be wrong?
Yes. A 60% estimate leaves 40% probability for the other outcomes.
Why keep yesterday's original percentages?
They allow users to evaluate what the model actually estimated before the result was known.
Should a goal in first-half stoppage time count?
Yes. If the goal is valid and occurs before the referee ends the first period, it forms part of the Half Time score.
Where can I see today's Half Time predictions?
Visit Half Time Predictions Today.
When the First Half Officially Ends
A standard football match consists of two playing periods, with additional time allowed for time lost. The Half Time result is therefore the score when the referee brings the first period to an end, including the applicable added time.
For the official framework, see IFAB Law 7 – The Duration of the Match.
This distinction is important for historical prediction settlement because a valid goal scored during first-half additional time can change the interval result immediately before half-time.
Historical Half Time Results Do Not Guarantee Future Predictions
Betiball provides football statistics and probability estimates rather than guaranteed outcomes. Yesterday's settled predictions can help users examine historical performance and model calibration, but previous accuracy cannot determine what will happen in future first halves.
Learn more about Betiball through About Us, FAQ, Terms of Use, Privacy Policy, Avoid Scam and Contact Us.
If you choose to gamble, only do so where legal and if you meet the applicable age requirements. Historical prediction performance should never be treated as a guarantee of future results. See Responsible Gaming for further information.