Opinion

The Board of Prison Terms (Board) appeals the order partially granting Wendell Holmes’s petition for habeas corpus. Although we dismiss the appeal as moot, we discuss part of the order requiring the Board to determine why Holmes should not be released within one year of his parole suitability finding or to give specific reasons why his release date should be beyond that one-year time.1 We explain why we have concluded the court incorrectly applied ex post facto principles and accordingly erred in issuing the order.

I

Holmes was convicted in 1975 of first degree murder and sentenced in that year to life imprisonment. In 1983, the Board held two hearings *485concerning Holmes’s suitability for release, one under the determinate sentencing law (DSL) and one under the indeterminate sentencing law (ISL) parole board rules (PRB) promulgated in 1976 as required by In re Stan-worth (1982) 33 Cal.3d 176 [187 Cal.Rptr. 783, 654 P.2d 1311]. Both panels found Holmes suitable for parole. The DSL panel set a parole date of June 20, 1992; the ISL panel set a parole date of April 20, 1990. Both panels used regulations promulgated after the date of Holmes’s crime. Contending the use of such regulations was unlawful, Holmes sought habeas corpus relief in the superior court in 1988.

After appointing counsel and holding extensive hearings, the court found uncontradicted evidence established the “Parole Board had a rule by practice that the vast majority of first degree murder life prisoners were paroled within one year of being found parole suitable at the time Holmes’ crime was committed.” Based on this finding and the fact that all such prisoners had been released within 34 months of being found parole suitable, the court ruled Holmes was entitled to be released unless the Board could give specific reasons why his release date should be beyond that one-year time. The Board has appealed this order. Pending the appeal, we are told the court-ordered hearings have been held and Holmes has now been released.

Discussion

The court made its ruling on the assumption that ex post facto principles required the Board to treat Holmes in the same manner as it had treated other first degree murderers who were serving a life sentence at about the time Holmes committed his crime. The court’s premise is valid, however, only if the retrospective application of the regulations here disadvantaged Holmes. (In re Seabock (1983) 140 Cal.App.3d 29 [189 Cal.Rptr. 310].) Not only must the application of the regulations be retrospective, i.e., apply to events occurring before its enactment, but they must disadvantage the offender affected by them. (Weaver v. Graham (1981) 450 U.S. 24, 29 [67 L.Ed.2d 17, 23, 101 S.Ct. 960].) Thus the crucial question before the trial court was not merely whether the later parole regulations were applied retrospectively, but whether their application “disadvantaged” Holmes.2

There are undoubtedly many events which occur during a prisoner’s sentence which the prisoner thinks are important. For example here *486Holmes certainly thought the Board’s determination that he was suitable for parole was significant. Our analytical ex post facto perspective is necessarily different than that of Holmes. The date on which parole suitability is found is irrelevant for ex post facto purposes unless it impacts on the ultimate release date of the prisoner. To validly compare punishment between two separate groups of prisoners the entire period of incarceration must be compared, not merely a part of that period. The fact that different points in time during a prisoner’s incarceration may have been selected to determine whether the prisoner is suitable for parole does not necessarily mean that the prisoner who remains incarcerated for a longer period of time following a parole suitability finding will be treated more severely than the prisoner who was released after a shorter period. The respective sizes of two baskets holding apples and oranges can not be compared by merely counting the apples in each. One must also count the oranges. Similarly here both components of the prisoner’s sentence must be counted—the period before the finding of parole suitability as well as the period following that date.

The focus of the trial court’s analysis should have been on Holmes’s loss of freedom, i.e., the period commencing with his commitment to prison and ultimately his release on parole. There is no detriment to Holmes and those similarly situated first degree murderers if such persons remain incarcerated for essentially the same period of time as first degree murderers who had received the benefit of the Board’s earlier practice of release within one year of a finding of parole suitability.

In this case the record shows Holmes did not remain in prison any longer than the comparable group of prisoners released within one year of the Board’s finding of parole suitability. In any event, because the trial court here limited its finding to the disparate dates of release between Holmes and the group of prisoners to whom he was compared following the Board’s finding of parole suitability, the trial court failed to make the essential inquiry of whether Holmes was disadvantaged by the ex post facto application of the 1983 regulation. Lacking the factual predicate to support its conclusion, the court prejudicially erred in ordering the Board to hold an additional hearing to explain why Holmes was not released within one year of his parole suitability date.

*487Disposition

Appeal dismissed.

Todd, J., concurred.

Benke, J., concurred in the result.

A petition for a rehearing was denied October 19, 1989, and petitioner’s application for review by the Supreme Court was denied January 3, 1990.