People v. State

fairly undermining public confidence in the administration of justice

Prosecutors should never lose.

March 28, 2011 By: John Kindley Category: Gerry Spence, Prosecutors, Vincent Bugliosi

In a comment on Mark Bennett’s post criticizing a Colorado district attorney’s plan to offer cash bonuses to her deputy prosecutors who participate in at least 5 trials in a year and win a felony conviction in at least 70% of them, I remarked:

I agree with Gerry Spence: If a prosecutor is doing his job right, he should never lose at trial.

[As I noted in this post, Spence wrote of his stint as a prosecutor in The Making of a Country Lawyer:

I finished my second term having tried many more cases, none of which I lost, not that such a record stands for much. With all the power prosecutors possess, they ought not lose cases. The wrong case, the unjust case should be rejected in the prosecutor’s office before he seeks an indictment.]

Mark replied:

I used to think that was right, but no longer.

Just as sometimes the defense has to try cases that are probably losers (and no criminal-defense lawyer worth his salt has never lost a jury trial), sometimes the State has to try cases that could go either way, and let the jury decide.

A prosecutor could easily bat 1.000 by selling the hard cases cheap. But that’s neither good lawyering nor, I suspect, justice.

Harris County (Houston, TX) Criminal Court No. 6 Judge Larry Standley commented:

RE: ” Gerry Spence: If a prosecutor is doing his job right, he should never lose at trial.” I simply do not understand this concept: Define “winning” (no Charlie Sheen quips please). Define losing.

As a former prosecutor for 14 years I received many “not guilty” verdicts that were actually “not proven beyond a reasonable doubt verdicts. No war story beating of my chest here – maybe a “better prosecutor” could have carried the day – but I didn’t and I still to this day feel “I did my job right”.

I dismissed and received no bills on many cases that some prosecutors might see as losing – but it was the right thing to do. My personal definition of justice is an abstract goal to strive for, but must be tempered by “The Rules”. In short it is this: “Justice: That state of affairs that exists when all people get what’s coming to them”.

In a simple black and white comparative example: For the innocent, justice would be never being arrested, charged or in the alternative a dismissal or not guilty. To the guilty, with all rules being followed by all parties in the C.J. System, that would be an arrest, charge and conviction.

Though imperfect, we still do have the BEST Criminal Justice System in the world. The best reward ANY prosecutor should ever hope for in this system is – at the end of the day, after doing their best, and following the rules – is that particular outcome of any particular case was the most “JUST” under the specific circumstances of that particular case give all the admissible evidence presented. In short – this concept of truly seeking justice ALONE, while doing one’s best, is the best and honest reward any prosecutor should ever need……..because we know it “ain’t for the money”.

I replied to Judge Standley:

Suppose a prosecutor is 80% sure, based on the evidence, that the defendant is guilty. Is it ethical for him to proceed to trial? I don’t think so, because he knows very well that 80% certainty is enough for many juries to convict. But if many juries are convicting based on 80% certainty, that means a whole lot of innocent defendants are being convicted. By regularly proceeding to trial in such circumstances, the prosecutor is guaranteeing that he’ll lose some trials that he should lose and — much worse — that he’ll win some trials that he should lose.

Suppose a prosecutor is 95% sure that the defendant is guilty — which is a little closer to what I think proof beyond a reasonable doubt requires. To my mind, a 5% doubt is still a reasonable doubt, and the prosecutor should [if he doesn’t dismiss the case outright] make an offer which would make it unreasonable for the defendant to proceed to trial. Now, in such circumstances you’ll sometimes have defendants who, either because they’re in fact innocent despite evidence showing it’s 95% likely they’re guilty or because they’re both guilty and unreasonable, refuse all reasonable offers and insist on proceeding to trial. Such a case might result in a guilty verdict. It might result in a hung jury. But if you get 12 jurors agreeing unanimously that the State didn’t prove its case beyond a reasonable doubt, especially given that jurors are notoriously not excessively fastidious about convicting people the State asks them to convict, I don’t see how that’s not a true black mark on the prosecutor’s record, and doesn’t call into question either the prosecutor’s trial skills or his wisdom in proceeding to trial in the first place. Prosecutors get to choose their cases, and there’s no shame in choosing them so that they virtually never lose. I’ve had the pleasure of visiting with the jury in the jury room along with the prosecutor and the judge after the jury acquitted my client in less than 30 minutes, and hearing them pointedly ask the prosecutor in so many words whose bright idea it was to prosecute the case. Although I was happy and relieved to have won (I knew juries are capable of anything), there was no hand-shaking going on between me and the prosecutor. I was glad to see the jurors applying salt to the prosecutor’s fresh wound, to see them adding insult to [the prosecutor’s failed attempt to cause] injury.

Granted, a prosecutor who only proceeds to trial in cases he’s almost certain to win, and who offers very reasonable plea offers to defendants who are 95% likely to be guilty, risks being seen as insufficiently aggressive by the electorate. Such a prosecutor might also be unpopular among defense attorneys who like to occasionally try beatable cases. But it seems an unbeaten trial record would also be a nice selling point at re-election time, along with the savings to the county in resolving most cases without trial.

Even in slam-dunk cases the prosecutor can and should offer the defendant something as consideration for saving him and the county the trouble of trial, e.g., a cap on the executed portion of the sentence that is less than the maximum the judge could reasonably impose after the defendant is convicted at trial. The prosecutor is truly abusing his discretion if his intent is to force a slam-dunk case to trial so as to get more trial experience and/or to improve his trial win-loss record.

BTW, I’ve come up with a definition of Justice that is similar to yours: Justice is the absence of crime. The “justice” that is doled out by the criminal “justice” system is Justice only in a derivative sense, i.e., only insofar as by deterrence, rehabilitation, [incapacitation] and/or (more controversially) retribution it tends towards or (more controversially) approximates the “absence of crime.” It is better that 100 guilty men escape than that 1 innocent suffer. The conviction of an innocent man is itself a crime. The escape of a guilty man is not.

I don’t mean to say by my comments above that a prosecutor who loses a trial is ipso facto a “bad” prosecutor. Marcia Clark and Christopher Darden, e.g., were of course right to try O.J. Simpson for murder, given the evidence against him. Although Vincent Bugliosi, among others, was highly critical of their performance at trial and argued that this is why Simpson got away with murder, it’s possible that Clark and Darden could have done everything right and still lost. Nevertheless, for the reasons stated above, such losses should be exceedingly rare. Bugliosi himself won 105 out of 106 felony jury trials during his career as a prosecutor. This conviction rate at trial was presumably due not only to Bugliosi’s ability and hard work but also to his discretion in deciding which cases to prosecute and what plea offers to make.

10 Comments to “Prosecutors should never lose.”

  1. David Schwartz says:

    Suppose a prosecutor is 100% convinced that a person is guilty (say, due to evidence that was excluded), but only 50% convinced that he can prove it in court beyond a reasonable doubt. You also seem to assume that Defendants will be rational in accepting a plea deal and that juries will be rational in convicting the guilty.

    • John Kindley says:

      I had thought of your scenario involving excluded evidence but it didn’t make it into my post. The jury has mightily failed in its duty if it convicts where the admissible evidence barely reaches the civil preponderance of the evidence standard. The prosecution should lose, but would be asking the jury to convict anyway. Hopefully the court of appeals recognizes that given the admissible evidence no rational juror could have been convinced beyond a reasonable doubt of the defendant’s guilt. But yeah, if the prosecution knows the defendant committed a real crime based on excluded evidence the jury won’t be presented with, and loses (or wins) at trial, that’s not as shameful or unjustifiable as losing (or winning) a case where the defendant’s guilt is as doubtful to the prosecutor as it is to everyone else. Ideally, though, the prosecutor would disregard inadmissible evidence in making trial and plea offer decisions, just as that evidence will be disregarded at trial.

      I don’t assume Defendants will be rational in accepting a plea deal. As I indicated in the post, the irrationality (or innocence) of a Defendant might lead to trial and acquittal where the defendant’s guilt seems almost but not quite certain and the defendant refuses a reasonable plea offer. I also don’t assume that juries will be rational in convicting the guilty, as my reference to the O.J. Simpson trial was meant to admit.

      I overstated the case a bit by saying prosecutors should “never” lose. I tried to moderate that some in my final paragraph by saying they should almost never lose. Spence and Bugliosi managed that feat. While not every prosecutor can be a Spence or a Bugliosi, I’d suggest a prosecutor is doing something wrong if he’s only winning 70% of his trials.

      I also used the word “should” in a deliberately weaselly way. If a prosecutor has actually proved his case beyond a reasonable doubt, then he “should” not lose that case, even though he in fact might lose that case.

      • David Schwartz says:

        Okay, let me make my point more forcefully. Suppose a prosecutor believes that the person he is charging is guilty beyond a reasonable doubt and believes that he has a 50% of convincing a jury that the person is guilty beyond a reasonable doubt. Are you saying he should not prosecute this case?

        In other words, I’m saying that you are essentially arguing that a Prosecutor should have an internal standard for prosecution that is much higher than “beyond a reasonable doubt”. Because if his personal standard is whether he believes he has, say, a 50% chance of convincing 12 jurors of guilty beyond a reasonable doubt, he’ll lose about half the time.

        I would argue that if a Prosecutor believes personally that the person is guilty beyond a reasonable doubt and the Prosecutor believes he has a 50% chance of convincing a unanimous jury of this, he should prosecute. And if he does so, he will lose about half of these cases.

  2. John Kindley says:

    I’d estimate that if the objective evidence showed an 80% likelihood that the defendant is guilty, a prosecutor probably does have around a 50% chance of convincing 12 jurors to convict him, which you argue should be sufficient for the prosecutor to prosecute. But an 80% likelihood is not “beyond a reasonable doubt,” and a prosecutor who is satisfied by it is likely to convict a lot of innocent people in the course of losing a lot of trials. So I’m not suggesting that a prosecutor should have an “internal standard” for prosecution that is higher than “beyond a reasonable doubt,” but that he should personally be so convinced by the admissible evidence of the defendant’s guilt that he should be around 95% confident that a jury will convict based on that evidence. That is, he should rationally believe that only an irrational jury would fail to convict based on the evidence. It seems to me that being persuaded “beyond a reasonable doubt” requires nothing less.

    • David Schwartz says:

      So, if he’s personally convinced beyond a reasonable doubt and believes the evidence is 50% likely to convince a unanimous jury, should he prosecute or not?

      My hypothetical was never about someone who the evidence showed was 80% likely to be guilty. Of course you should never prosecute such a person. All of my hypotheticals were about people the Prosecutor was firmly personally convinced were guilty beyond a reasonable doubt.

      Again, this is my point: “I would argue that if a Prosecutor believes personally that the person is guilty beyond a reasonable doubt and the Prosecutor believes he has a 50% chance of convincing a unanimous jury of this, he should prosecute. And if he does so, he will lose about half of these cases.”

      • John Kindley says:

        And my counterpoint was that if based on the admissible evidence the prosecutor can only be 50% confident that a jury will convict then that evidence does not objectively constitute “proof beyond a reasonable doubt,” and the prosecutor should not prosecute. (If he’s more than 50% confident of his chances at trial, then he should probably prosecute but offer a plea that would make it unreasonable for the defendant to take his chances at trial.) The only scenario in which a rational prosecutor could be personally convinced beyond a reasonable doubt that the defendant is guilty and yet still only be 50% confident of winning at trial would be the scenario, addressed above, in which the prosecutor’s belief in the defendant’s guilt is based on evidence which will be inadmissible at trial.

  3. David Schwartz says:

    Ah, so at root we have a factual dispute. You don’t think it’s at all common for a Prosecutor to be justifiably convinced that someone is guilty beyond a reasonable doubt but also believe that a jury is not much more than 50% likely to convict.

    Perhaps we can at least agree that a reasonable Prosecutor in such a case should prosecute. And perhaps we can also agree that a reasonable Prosecutor will lose a fair percentage of such cases.

  4. John Kindley says:

    I can agree that a prosecutor who is justifiably convinced beyond a reasonable doubt of a defendant’s guilt should in general prosecute — although some imaginable scenarios involving very weak cases after important evidence is excluded remain problematic, since the prosecutor would essentially be giving the jury an opportunity to fail in its duty to acquit absent proof beyond a reasonable doubt and hoping it does so. The key words to me though are “justifiably” and “reasonable.” Prosecutors should be reality-based, and their personal and notion of what “beyond a reasonable doubt” requires should not be driven by a “close enough for government work” philosophy or the idea that their job to lock up as many people as they can. Rather, it seems to me that, because they’re public servants, their personal notion of what “beyond a reasonable doubt” requires should be closely correlated to their assessment of the likelihood of obtaining a conviction from a jury drawn from the public the prosecutor serves. Granted, some communities might be inclined to vote for a prosecutor who promises to try to lock up as many people as he can. For my part, I’d be more inclined to vote for the prosecutor who promises to try to resolve as many cases as he can by plea and who wins the vast majority of the trials he’s not able to plea bargain out.

  5. suppose a prosecutor hears an accusaton from an 18 year old girl and never does any investigation whatsoever, does not take a wrotten statement form the na, 23 days later out of memory makes a report, and has the man arrested> he makes that man wait 3 1/2 years to go to trial and all the while recieves many many investigative reports from the defendants attorney stating the accuser was a liar and he not one time does sany type of follow up on those reports or the people who gave them, no investigation or even speak to people who were involved in the lives, home, neigborhood. he calls in the alledged victim 1 week beofre the rial to show her the reports and the proof she lied several times on her intitial report that was also used tyo have the man indicted and takes the man to trial all the while stating openly he cannot win, she he be held accountable????

  6. What do we have when, to secure a wrongful conviction, a senior assistant public defender helps and actually mentors a junior assistant states attorney?


2 Trackbacks/Pingbacks

  1. Reasonable Doubts About Dominique Strauss-Kahn And His Prosecution | Popehat 23 08 11
  2. Prosecutors are not your friends. | People v. State 26 08 11

Leave a Reply


  • "[T]here is just nothing wrong with telling the American people the truth." - Allen v. United States

  • Lysander Spooner

    Henry George

    Harriet Tubman

    Sitting Bull

    Angelus Silesius

    Smedley Butler

    Rose Wilder Lane

    Albert Jay Nock

    Dora Marsden

    Leo Tolstoy

    Henry David Thoreau

    John Brown

    Karl Hess

    Levi Coffin

    Max Stirner

    Dorothy Day

    Ernst Jünger

    Thomas Paine