To know what to do at any point in a jury trial, we must have some theory of how the parts of the trial fit together.
In a felony jury trial with twelve jurors (and the requirement of a unanimous verdict) there are 4,095 permutations of verdicts that do not involve our client going to prison. ((Each juror’s individual verdict can be either “guilty” or “not guilty.” The combined verdict of the entire jury can be expressed as a 12-digit binary number, with each bit representing one juror’s vote, 0 for guilty and 1 for not guilty. There are 212 = 4,096 12-digit binary numbers. Only one of them, 00000000000, or 0, is a unanimous guilty verdict.)) There is only one permutation that convicts our client. If each juror’s verdict were independent of the others and there were a 50% chance that each juror would convict, there would be a 1 in 4,096 chance of a conviction.
But jurors’ verdicts are not independent—one juror’s verdict will affect another’s—and the chance that any given juror will acquit is not necessarily 50%.
Our goal as criminal-defense lawyers is to keep the government as far as possible from that one permutation that convicts our client. To do this we need to recognize how jurors’ verdicts are dependent on each other, and how each juror reaches a verdict.
Many trial lawyers believe that every trial is won or lost in jury selection. Modern research into human cognition suggests that once a person forms a belief, it is very difficult to change that belief. Mere evidence will not suffice to change a normal person’s belief; she will disregard evidence that contradicts her beliefs and will magnify the evidence that supports her beliefs. This is called confirmation bias. Other biases act toward the same goal: to lock a person into her initial belief.
While we make jurors promise to consider all of the evidence before reaching a decision, there is very little evidence that they actually do this.
Research with mock juries suggests that how the jury is split when it goes out predicts the verdict. A jury that is 7–5 for conviction on the first ballot will most likely convict or hang; a jury that is 11–1 for conviction on the first ballot will almost certainly hang. A jury that is 6–6 or better for the defense on the first ballot will likely not convict. It appears that it is force of personality, rather than quality of evidence, that determines whether jurors will change their minds.
So we want to have as many jurors as possible on our side when they go out (we don’t want our defense to depend on beliefs formed in the jury room), which means that we want to have as many jurors as possible on our side as early as possible.
How early is “as early as possible”? Most jurors form a belief about the right result in the case by the end of opening statement; this belief will not be changed absent blockbuster evidence that they have not been primed to expect.