An ideal model should have low bias and low variance. In that case, all the predictions will be near to the bulls-eye.
In the case of low bias and high variance, even though the model predictions will be close to the bull’s eye, it will have some variability meaning that it can perform well for one dataset, but it may also perform terribly for another dataset.
In case of low variance and high bias, the model will not be close to the bulls-eye, but they will be concentrated in a single location, as it has low variance.
In case of high variance and high bias, the model will be both not close to the bulls-eye and also it will not be concentrated in a single location.