A 1X2 forecast gives three probabilities: home win, draw and away win. They add up to 100%. Illustrative example (fictional, not a WinSight forecast): Home 50%, Draw 27%, Away 23%. This does not say the home side "will win". It says that across many matches rated this way, the home team should win about half of them. The other half of the time it won't — and that is the model working as intended, not failing. So judge a forecaster over hundreds of matches, not one weekend. Our track record (/accuracy) settles every published call, misses included.
FAQ: Is a 60% favourite a "safe" pick? No. It is expected to fail four times in ten.