The line in Las Vegas shops on Super Bowl XIX opened with San Francisco favored by 2 or 2 1/2 or 3.

"There's more 3 than 2 1/2," Las Vegas odds maker Bob Martin declared, "But by the end of this week, I have a feeling it will come down to 2 1/2 in most places. There's going to be a little more money showing for the Dolphins. It sometimes takes time to travel."

I'm picking the 49ers, having been careful to shop for the 2 1/2 instead of having to give 3 points. The risk is an imaginary $1,000 as, for the second straight year, I need to win the Super Bowl in order to atone for the sins of a season that included some poor money management.

Super Bowl XIX figures to provide another exciting championship contest inasmuch as it brings together the NFL's two best teams. The best teams don't always make it to the Big Game, of course. Lately, they have.

San Francisco has beautiful balance; Miami has a beautiful passer. The 49ers are 17-1; the Dolphins 16-2. San Francisco played one of the league's easiest schedules; Miami's opposition was only slightly tougher.

There are few areas in which Miami is superior to San Francisco. Punting (Reggie Roby is terrific) and pass protection are two, in addition to quarterback Dan Marino's brilliance. But such comparisons are meaningless. Offensive lines don't play each other. It is the matchups that matter. For instance, the fact that Miami's offensive line will, once again, provide Marino ample time to throw is a consideration that is hard to overcome.

Don Shula's preparation usually gives Miami an edge, but not when the rival is Bill Walsh and Walsh's team is playing near home, in Palo Alto, Calif., on the Stanford campus.

Certainly, Shula must be worried by the ineffectiveness of kicker Uwe von Schamann this season. Ray Wersching gives the 49ers a decided advantage in field goal accuracy.

The week prior to the conference championship games, when I was asked to pick the Super Bowl winner, I took Miami. But my worst fears concerning the Dolphins' defense were confirmed in Miami's defeat of Pittsburgh. That defense leaves a lot to be desired. Pittsburgh ran on Miami and, if the Steelers had a better passer, they would have put 40 points on the board.

I believe San Francisco will score at least 38 points Sunday, compared to Miami's 31 or more. Miami doesn't figure to stop San Francisco's rushing or passing. San Francisco will contain Miami's running, not that it matters. Marino will have another good day. He beat the Los Angeles Raiders' defense. He'll beat San Francisco's. But the 49ers will do what the Raiders did: outscore the Dolphins. They have the offense to do just that.

One or two fumbles by 49ers halfback Wendell Tyler could turn my projected result upside down. And quarterback Joe Montana must be sharp, obviously. I don't think he has been as consistent this year as he is being portrayed, but he has a history of rising to the occasion. And the Dolphins' defense can be made to look soft whereas Marino will be playing against a defense that has no weakness.

San Francisco should be able to control the ball enough to keep Marino on the bench longer than he's accustomed. That's the edge. I also have a hunch San Francisco will be more effective scoring from inside the 10. Give the 2 1/2 or 3.

For those fans who need more action on Super Bowl XIX, Martin spent last weekend thinking up a special card for the Union Plaza. All a bettor has to do is to make 15 winning picks (ties, you lose) from a list that included the following:

* 49ers giving 3 or Miami getting 3. (I took San Francisco.)

* Winner of each quarter. (I took San Francisco in the first and fourth, Miami in the second and third.)

* Total points, 54. (I went over instead of under.)

* Total points first half, 29. (I went over.)

* Total points by quarter, over or under, 10, 14, 13, 14. (I went over, over, over, over.)

* Marino to pass for 85 more yards than Montana. (I took Marino for at least 86 more yards.)

* Total field goals, three. (I took more.)

* Wersching to kick one field goal more than von Schamann. (I took Wersching for at least two more.)

* Mark Clayton to catch one pass more than Dwight Clark. (I went under, with Clark.)

* Marino to pass for three touchdowns. (I took him for four or more.)

* Tyler to fumble 1 1/2 times. (I took one or fewer, with my fingers crossed.)

"You pick any 15 of these correctly, going 15 for 15, and the reward is $125,000 for $5 or $250,000 for a $10 bet," Martin said. "A box at the bottom of the card has a space for the contestant to write in the name of the player he thinks will score first. If you pick 15 for 15 (remember, ties lose) and you pick the player who scores first (I took Clark, Wersching was too obvious), you'll collect double, or $500,000 on the $10 bet."

Anybody lucky enough to do that should get $500 million in a fair world. The ties alone are guaranteed to kill you. But it's fun trying. Gimmick betting is forever popular and suckers for such football cards -- even those designed in Las Vegas during Super Bowl week -- are born every other minute. THE RECORD

Won-lost record: 46-41 Net for Week Net for Season -- $75 -- $925

Jan. 6 results: Miami, giving 10, defeated Pittsburgh, 45-28, plus $750; Chicago, getting 10, lost to San Francisco, 23-0, minus $825.