Times/Siena surveyed the same group of voters pre and post conviction. They saw a 2 point shift in the race after the conviction.
The true shift is more like 3 points if you adjust for non-response of people from the pre-conviction sample.
This early this is about the best way to measure the effect of the conviction as it avoids the sampling error of surveying a different group of voters.
Notably, those who dislike both Biden and Trump were especially likely to shift away from Trump after the conviction.
Also low information voters were especially likely to shift as well. I keep reminding people there are a large chunk of voters that really don't follow the news and politics that are particularly up for grabs with events like this.
Some Trump supporters shifted to Biden following the conviction, so it wasn't all from undecided voters.
46% approve of the verdict. Only 33% disapprove of the verdict.