After a First-Leg Draw in Handball: Who Goes Through? A Model
A level first leg feels like a reset. In this model it is not: the side hosting the return is the favourite, and the size of that edge depends on two inputs.
Written by Lars MikkelsenHandball analystIn a stated handball scoring model, after the first leg of a two-legged tie ends level, how often does the side hosting leg 2 go through, and when does the draw's score matter?
Model study, no match records: exact probability calculation from a stated possession model of handball (about 55 possessions per side per 60 minutes, 53% of possessions scored)
- 164.8%leg-2 host goes through after a draw, equal teams
- 21.54fair odds, leg-2 host to go through
- 37.0%chance leg 2 is also level (shootout)
- 435.2%host goes through even when 4 goals weaker
Model studyEvery number on this page is calculated from the rules of the sport and stated assumptions. These are not historical match results.
A first leg ends level and the commentary calls the tie wide open. I wanted to test that with a stated possession model of handball: about 55 possessions per side per 60 minutes, with 53% of them scored. The model says that between equal teams the side hosting leg 2 goes through 64.8% of the time, fair odds of 1.54.
That edge comes entirely from an assumed 2-goal home advantage, which is the main limitation of the whole study. Every figure here is a probability calculation, not a match record. The model has no goalkeeper hot streaks, no switch from 6-0 to 5-1 at half-time, no passive-play calls and no late time-wasting, and its home edge is an input rather than something measured. If you want the match-by-match view, my today's handball tips page carries it. This piece is about the structure underneath.
Aftermath ยท 01The leg-2 host's chance across true strength gaps
Read this chart left to right as the host getting stronger. Equal teams sit at 64.8%. A host that is truly 4 goals stronger goes through 87.4% of the time, and 6 goals stronger reaches 93.7%. The striking end is the weak side. A host that is 4 goals weaker still goes through 35.2% of the time, and even a 6-goal-weaker host manages 22.3%.
View the data: leg-2 host goes through after a draw
| Model, aggregate only, home edge 2 goals; host's true gap | |
|---|---|
| Host weaker by 6 | 22.3% |
| Host weaker by 4 | 35.2% |
| Host weaker by 2 | 50.0% |
| Equal teams | 64.8% |
| Host stronger by 2 | 77.7% |
| Host stronger by 4 | 87.4% |
| Host stronger by 6 | 93.7% |
For a bettor, the draw has not wiped the tie clean. It has handed one full match, at home, to the side hosting leg 2. In this model, the outsider hosting the return carries a real chance that a quick glance at the ratings can miss.
Aftermath ยท 02Home edge and strength gap trade off goal for goal
This heatmap is the one I would study before forming any view of a return leg. With no home edge, equal teams make the tie a coin flip at 50.0%. Raise the home edge to 4 goals and the same equal host goes through 77.7%. Each goal of home advantage moves the figure exactly as much as a goal of true strength, so the values run along the diagonals.
That is why I convert any price or rating into an expected goal margin first and then add my own view of the arena. A hostile hall with a goalkeeper who feeds off the crowd is a different input from a half-empty sports centre.
View the data: host's go-through chance: home edge ร strength
| โ4 | โ2 | 0 | +2 | +4 | |
|---|---|---|---|---|---|
| Home edge 0 | 22.3% | 35.2% | 50.0% | 64.8% | 77.7% |
| Home edge 1 | 28.4% | 42.4% | 57.6% | 71.6% | 83.0% |
| Home edge 2 | 35.2% | 50.0% | 64.8% | 77.7% | 87.4% |
| Home edge 3 | 42.4% | 57.6% | 71.6% | 83.0% | 91.0% |
| Home edge 4 | 50.0% | 64.8% | 77.7% | 87.4% | 93.7% |
If your home-edge figure is wrong by a goal, your read on the tie is wrong by a goal of strength. This heatmap shows how much that one input moves the answer.
Aftermath ยท 03Most ties are settled in regulation, not shootouts
Between equal teams, the host wins leg 2 outright 61.3% of the time and loses it 31.6%. Leg 2 also finishes level 7.0% of the time, and in this model that sends the tie to a 7-metre shootout treated as 50/50. When the host is 4 goals stronger it wins leg 2 outright 85.4% of the time. When it is 4 goals weaker, that falls to 31.6%.
View the data: how the tie is settled after a first-leg draw
| Host wins leg 2 | Draw, host wins shootout | Draw, host loses shootout | Host loses leg 2 | |
|---|---|---|---|---|
| Host โ 4 | 31.6% | 3.5% | 3.5% | 61.3% |
| Host โ 2 | 46.2% | 3.8% | 3.8% | 46.2% |
| Equal teams | 61.3% | 3.5% | 3.5% | 31.6% |
| Host + 2 | 74.9% | 2.8% | 2.8% | 19.4% |
| Host + 4 | 85.4% | 2.0% | 2.0% | 10.6% |
The shootout slice is thin in every row. In this model, backing either side to go through is close to backing them to win the second match. The 7-metre lottery is a small extra, not the main event.
Aftermath ยท 04Under away goals, the draw's scoreline shifts the odds
Under aggregate only, the score of the draw is irrelevant: equal teams sit at 64.8% whether the first leg ended low or high. Add an away-goals rule and that changes. A 22-22 first leg cuts the equal host's chance to 61.4%, and 26-26 gives 62.4%. A 30-30 draw lifts it to 65.7%, and 34-34 to 68.0%. The last case is high because the visitor would need 34 or more in a level leg 2 to go through.
Away-goals rule: a level aggregate is decided by away goals, then a 50/50 shootout if those are level too.
View the data: does the first-leg score of a draw matter?
| Aggregate + shootout | Away-goals rule | |
|---|---|---|
| 22-22 first leg (equal teams) | 64.8% | 61.4% |
| 26-26 first leg (equal teams) | 64.8% | 62.4% |
| 30-30 first leg (equal teams) | 64.8% | 65.7% |
| 34-34 first leg (equal teams) | 64.8% | 68.0% |
The logic runs both ways. In a low-scoring draw, the leg-2 host banked only a few away goals, so a tight, defence-heavy return leg can favour the visitor on that tiebreak. In a shoot-out first leg, the host holds a large store of away goals that the visitor has to match.
What happens next
- Price the host as favourite
Between equal teams the model gives the leg-2 host 64.8%, fair odds of 1.54. Treat any price near even money on the host after a level first leg with suspicion, not as a gift to the visitor.
- Respect the weaker host
A host that is truly 4 goals weaker still goes through 35.2% of the time in this model. Before you dismiss the home outsider, convert its rating into an expected goal margin.
- Check the tiebreak rules first
Under away goals, a 22-22 draw drops the equal host to 61.4%, while 34-34 lifts it to 68.0%. Under aggregate only the scoreline is irrelevant, so confirm which rule applies before you read anything into the first-leg score.
Questions bettors ask
Who usually goes through after a first-leg draw in handball?
In this model the side hosting leg 2 goes through 64.8% of the time between equal teams, because of an assumed 2-goal home edge. With no home edge, the model makes it a coin flip at 50.0%.
How often does a two-legged handball tie go to a 7-metre shootout after a draw?
Between equal teams, the model has leg 2 also finishing level 7.0% of the time. Under aggregate-only rules, that sends the tie to a shootout.
Does the score of a first-leg draw matter under away goals?
Yes, in this model. A 22-22 draw gives the equal-strength leg-2 host 61.4%, while a 34-34 draw gives 68.0%. Under aggregate only, both stay at 64.8%.
I write handball previews with a focus on defensive systems, goalkeeper form, and the second-half fatigue patterns that swing more matches than people realise.
View profile โ
