Purpose. Firesetters have traditionally been considered dangerous repeat offenders. However, the specific risk factors associated with firesetting recidivism have not been consistently tested in representative samples. It is also unclear whether individuals whose offending is limited to firesetting are at increased risk of reoffending when compared with firesetters who have more versatile offending. This study aimed to: (1) determine the rate of firesetting recidivism in a representative sample of firesetters before the courts; and (2) determine the psychiatric and criminogenic factors that are related to firesetting recidivism. Methods. The study employed a data linkage approach to examine the psychiatric and criminal histories of 1052 firesetters convicted of arson between 2000 and 2009 in Victoria, Australia. The characteristics of those who reoffended, over a follow-up period of 2.5-11 years, by committing arson and arson-related offences were compared with those who went on to reoffend in other ways but not arson. An improper model was used to determine which of the tested variables could meaningfully predict firesetting recidivism. Results. The rate of firesetting recidivism, based on charges, was very low (5.3 ) compared with the rate of general recidivism (55.4 ); the vast majority of firesetting recidivists were mixed (criminally versatile) offenders (91 ). The study found that general criminality, firesetting history, and psychiatric disorder were associated with firesetting recidivism. Conclusions. When assessing risk of firesetting recidivism, clinicians need to consider general criminality in addition to fire-specific history, and the potential impacts of mental disorder on recidivism.