Match ☆ 17 Dec 2026 · 20:00 UTC
ITL · UEFA Conference League · 2026 League phase
ITA Atalanta BC
Current form · last 20
L D W W L W W W L L L W W L L D D L D W
Latest results · oldest first
Current league place
#12 / 20 · ITA · Serie A · 2026/27
As of this morning · domestic league only
–:–
Scheduled
SWE Mjällby AIF
Current form · last 20
L D L L W L L L D D L D W L D L W D W D
Latest results · oldest first
Current league place
Current league standing unavailable
As of this morning · domestic league only
View head-to-head record →
The models forecast for this fixture
Draw predictor · this site's own figures
Figure Atalanta BC Mjällby AIF
N — no-draw runmatches in a row without a draw 1 0
Max Nthis team's longest run ever 16 14
E — draw debtexpected draws over the current run 0.26 0.00
p50 — last 50draw rate over its last 50 matches 24.0% 28.0%
p100 — last 100draw rate over its last 100 matches 27.0% 26.0%
Draw % — all timeshare of every match drawn 28.0% 24.6%
Draw scorethe predictor ranking score 0.119 0.078
Rankposition in that ranking 1,357 1,710
Σ Nboth sides' runs added 1
Σ Eboth sides' draw debt added 0.26
Σ Scorewhat the Cross predictor orders by 0.198
Team strength · Elo
Atalanta BC 1,883
Mjällby AIF 1,783
Rating gap +166 home advantage included
Draws at this gap 23.7% 150–199 · 53,477 matches
Goals model · Dixon-Coles
Expected goals 1.63 – 1.15 home – away
Home win 47.8%
Draw 25.7% every level score added up
Away win 26.4%
Both together · the blend
Blended draw chance 25.2% 75% goals model · 25% Elo gap
Goals model said 25.7% form only
This Elo gap said 23.7% strength only

The mix is the one that scored best over 544,576 matches — the two are averaged on the log-odds scale, not simply added.

The regression · the logistic model
Logistic draw chance 22.9% five signals, then recalibrated
95% band 22.9% – 22.9% how unsure the fit is, not the match
The blend says 25.2% the older published reading

A regression over five signals — the goals model, the draw rate at this Elo gap, the expected goals in the fixture and the competition's own base draw rate — with an isotonic map on top to put the level back where reality is. Held out on a decade it never saw it beat the blend above, which is why both are shown here while one is chosen. This is a real forecast: the fixture had not been played when the model was fitted.

A forecast from every match played so far. Measured over 160,869 priced matches the goals model is slightly worse than the bookmakers (how it is scored), so a high draw chance means an even, low-scoring match rather than value. Descriptive statistics, not betting advice.