If we took your principle which you are applying to criminals, it would also apply to non-criminals. There are any number of people whom we can guess are likely to commit violent crime at a later date. We can’t arrest them until there is some actual evidence that they really intend to do so.
The death penalty is a punishment for a specific act.
The prevention of future murders is a moral argument in support of the death penalty. It is not the reason for the execution.
FYI:
“The results are boldly clear: executions deter murders and murder rates increase substantially during moratoriums.”
(2003) Emory University Economics Department Chairman Hashem Dezhbakhsh and Emory Professors Paul Rubin and Joanna Shepherd state that “our results suggest that capital punishment has a strong deterrent effect. An increase in any of the probabilities – arrest, sentencing or execution – tends to reduce the crime rate. In particular, each execution results, on average, in eighteen fewer murders – with a margin of error of plus or minus 10.” (1) Their data base used nationwide data from 3,054 US counties from 1977-1996.
(2003) University of Colorado (Denver) Economics Department Chairman Naci Mocan and Graduate Assistant R. Kaj Gottings found “a statistically significant relationship between executions, pardons and homicide. Specifically each additional execution reduces homicides by 5 to 6, and three additional pardons (commutations) generate 1 to 1.5 additional murders.” Their "data set contains detailed information on the entire 6,143 death sentences between 1977 and 1997. (2)
(2001) University of Houston Professors Dale Cloninger and Roberto Marchesini, found that death penalty moratoriums contribute to more homicides. They found: “The (Texas) execution hiatus (in 1996), therefore, appears to have spared few, if any, condemned prisoners while the citizens of Texas experienced a net 90 (to as many as 150) additional innocent lives lost to homicide. Politicians contemplating moratoriums may wish to consider the possibility that a seemingly innocuous moratorium on executions could very well come at a heavy cost.” (3)
(2001) SUNY (Buffalo) Professor Liu finds that legalizing the death penalty not only adds capital punishment as a deterrent but also increases the marginal productivity of other deterrence measures in reducing murder rates. “Abolishing the death penalty not only gets rid of a valuable deterrent, it also decreases the deterrent effect of other punishments.” “The deterrent effects of the certainty and severity of punishments on murder are greater in retentionist (death penalty) states than in abolition (non death penalty) states.” (4)
(2003) Clemson U. Professor Shepherd found that each execution results, on average, in five fewer murders. Longer waits on death row reduce the deterrent effect. Therefore, recent legislation to shorten the time prior to execution should increase deterrence and thus save more innocent lives. Moratoriums and other delays should put more innocents at risk. In addition, capital punishment deters all kinds of murders, including crimes of passion and murders by intimates. Murders of both blacks and whites decrease after executions. (5)
(2003) FCC economist Dr. Paul Zimmerman finds: “Specifically, it is estimated that each state execution deters somewhere between 3 and 25 murders per year (14 being the average). Assuming that the value of human life is approximately $5 million {i.e. the average of the range estimates provided by Viscussi (1993)}, our estimates imply that society avoids losing approximately $70 million per year on average at the current rate of execution all else equal.” The study used state level data from 1978 to 1997 for all 50 states (excluding Washington D.C.). (6)
(2003) Emory University Economics Department Chairman Hashem Dezhbakhsh and Clemson U. Professor Shepherd found that “The results are boldly clear: executions deter murders and murder rates increase substantially during moratoriums. The results are consistent across before-and-after comparisons and regressions regardless of the data’s aggregation level, the time period, or the specific variable to measure executions.” (7)
dpinfo.com/death_penalty_as_a_deterrent.htm