The enormous problems with capital punishment—from wrongful convictions, wasted resources, botched executions, to troubling racial disparities—are becoming increasingly more objectively obvious. And states are recognizing this fact. In 2015, only 49 death sentences were issued in the entire United States (down from a peak of 315 in 1996), and only 28 convicts were executed (down from a peak of 98 in 1999).

The latest of the 19 states to completely end capital punishment (up from 12 in 2004) is Connecticut by virtue of a 5-2 decision of the Connecticut Supreme Court on May 27.

In September of 2006, I began an empirical evaluation of the operation of the Connecticut death penalty by looking at every homicide case that was capital eligible under Connecticut’s 1973 death penalty statute (the report, Capital Punishment in Connecticut, 1973-2007: A Comprehensive Evaluation from 4686 Murders to One Execution, first appeared in 2008 and was published in 2011). Of the 205 death eligible cases between 1973 and 2007, 12 received death sentences, of which three were later overturned.

My regression analysis showed that harsher treatment was given to cases in which minorities killed whites and that this difference could not be explained by legitimate factors such as the nature of the crime, the strength of the evidence, the judicial district in which it occurred, the number of victims, or other characteristics of the crime or defendant.

My findings stimulated intense political jockeying with many fits and starts, ultimately leading in April of 2012 to legislative abolition of the death penalty– albeit only prospectively. That set the stage for the dramatic 4-3 decision of the Connecticut Supreme Court in August 2015 holding that, after the Connecticut legislature had abolished the death penalty, it would no longer advance a legitimate penological interest to execute the remaining men on death row.

That historic ruling was put into limbo when a member of the court, the only African-American justice, retired right after the August decision and prosecutors in the state decided to simply ignore it and proceed towards executing the remaining death row inmates. Their hope was that the newly appointed justice would join the three dissenters to reverse the August decision.

Last week’s 5-2 decision by Connecticut’s Supreme Court ended this brazen ploy. The Chief Justice who had been insistent in the August decision that the death penalty should not be deemed unconstitutional understood that the reversal that the prosecutors sought would provide additional evidence of the arbitrariness underlying the operation of the Connecticut death penalty. Both she and the newly appointed Justice wrote that the considered decision of the entire court should not be reversed simply because one justice had been replaced.

The decision was clearly correct, in light of my research and relevant U.S. Supreme Court decisions. As the Court had stated in Atkins v. Virginia, unless the imposition of the death penalty “measurably contributes to one or both of these goals [deterrence and retribution], it ‘is nothing more than the purposeless and needless imposition of pain and suffering,’ and hence an unconstitutional punishment.” Although the Connecticut legislature’s prospective repeal of the death penalty eliminated any possible deterrence rationale, the dissenting justices sought to uphold the constitutionality of the death penalty on the grounds that killing those on death row would “measurably contribute” to retribution. They argued that the very small number of sustained Connecticut death sentences demonstrated that the death penalty was being appropriately limited to the worst of the worst offenders. But empirical data showed how hollow this claim was. Under any of the metrics that were developed to capture the egregiousness of the 205 death eligible cases, only one of the cases fell into the category of the worst of the worst offenses when judged without the corrupting influence of the race of the defendant and victim.

Connecticut will now stop wasting the millions it spent every year on extended death penalty trials and appeals, and hopefully will channel those resources to some beneficial criminal justice purpose. An obvious place to start is in solving murder cases. When Connecticut’s death penalty statute went into effect in 1973, the state was solving around 95 percent of its murders. Now more than 4 out of 10 murders are never solved, and the percentage of murders leading to arrests and convictions is even lower than this figure. Executing a handful of bad killers while thousands of murderers go free entirely serves neither deterrence nor retribution. Swift and certain punishment is the best and surest way to reduce crime and promote justice. This past week, Connecticut took an important initial step in that direction. I hope that the states still saddled with capital punishment regimes will learn from Connecticut and follow its lead.