IT PROVIDES EVALUATORS WITH (1) A MEANS OF EXAMINING SHORT-RUN PROGRAM IMPACTS ON THE POSTPONEMENT OF RECIDIVISM THROUGH ESTIMATES OF THE AVERAGE TIME AT WHICH RECIDIVISM OCCURS; (2) A WAY OF MEASURING LONG-RUN PROGRAM IMPACT ON THE PREVENTION OF RECIDIVISM THROUGH ESTIMATES OF THE ULTIMATE PROBABILITY OF RECIDIVISM; AND (3) A WAY TO DETERMINE IF INDIVIDUALS HAVE BEEN SUCCESSFUL LONG ENOUGH TO BE 'SAFE' USING ESTIMATES OF THEIR CONDITIONAL PROBABILITY OF FUTURE RECIDIVISM. THE STANDARD PROCEDURE FOR STUDYING RECIDIVISM IS TO DETERMINE THE PERCENTAGE OF INDIVIDUALS IN A COHORT WHO RELAPSE OR FAIL WITHIN A PRESCRIBED FOLLOWUP PERIOD OF GENERALLY 1 TO 3 YEARS. THIS METHOD, HOWEVER, IGNORES THE TIMING OF FAILURES AND MAY PROVIDE MISLEADING RESULTS. THE NEW MODEL POSTULATES THAT SIMILAR INDIVIDUALS HAVE THE SAME PROBABILITY OF FAILURE UPON RELEASE. BY CHANCE, SOME OF THESE INDIVIDUALS FAIL ALMOST IMMEDIATELY. THOSE WHO AVOID RECIDIVISM HAVE A SMALLER FUTURE PROBABILITY OF FAILURE; THE RATE OF FAILURE CONTINUES TO DECLINE AS THEIR SUCCESSFUL TIME AT RISK INCREASES. THIS PHENOMENON COULD BE CAUSED BY MANY FACTORS. FOR EXAMPLE, CONTINUED SUCCESS MIGHT INCREASE THESE INDIVIDUALS' CONFIDENCE TO THEIR ABILITY TO REMAIN SUCCESSFUL. IN ADDITION, IN THE CASE OF DRUGS, ALCOHOL, AND TOBACCO, THE LONGER SOMEONE HAS ABSTAINED, THE GREATER THE PERSON'S PHYSICAL AND SOCIAL ADAPTATION MIGHT BECOME. A MATHEMATICAL REPRESENTATION, DERIVATION, AND APPLICATION OF THE MODEL ARE PRESENTED, AS WELL AS A COMPARISON WITH THE MALTZ-MCCLEARY MODEL. IT IS CONCLUDED THAT ADDITIONAL ISSUES, POLICY IMPLICATIONS, AND OTHER MODELS SHOULD BE CONSIDERED IN FUTURE RESEARCH. TABULAR DATA, GRAPHS, REFERENCES, AND REFERENCE NOTES, ALONG WITH THREE APPENDIXES SHOWING THE COMPUTATION OF ESTIMATES ARE PROVIDED. (PRG)