Perhaps every baseball fan has a secret addiction. Mine, it's becoming obvious, is a statistic called "total average."

For the past two Aprils, this writer has been foisting his pet statistic on the public, touting it as an all-encompassing offensive yardstick that combines batting average, slugging average, on-base percentage and stolen-base percentage.

It might almost be a relief if someone demonstrated that "total average" was a fraud. Then my index finger wouldn't be numb from punching a pocket calculator. (A fitting punishment, no doubt, for anyone who creates yet another new baseball statistic.)

Total average is back. And shows no promise of going away.

It's still better than any other single statistic in baseball. TA may be a far cry from a supreme average, but its basic premise is so simple that it's hard to believe that its results aren't worth appraising.

This year's project includes:

A total average top 20 for each league.

A total average all-star team by position for each league.

An all-underrated list of part times who are total average stars.

And a list of aggregate base leaders: a stat that goes hand-in-hand with total average.

Gentlemen like Sixto Lezcano, Willie Stargell, Mike Ivie and Fred Lynn will be flattered by what follows. As for the likes of Jeff Burroughs, Lee May, Willie Montanez and Duane Kuiper (the least offensive man in baseball), it can only be hoped they curse the statistic and not its creator.

Basically, the premise for TA is that baseball has two fundamental units of measurement: the base and the out. The struggle -- in every inning -- is between bases and outs.

So let's create a statistic (analogous to the batting average) that measures the relationship between all the bases a player accumulates by his own offensive acts, and all the outs that he costs his team.

By all the bases, we mean all: including walks, steals and being hit by pitches. This brings batting, sluging, stealing and walking into play simultaneously and in proper proportion.

By all the outs, we mean all: including getting caught stealing, or creating an extra out by hitting into a double play.

To compute total average, give a player credit for each base he earns: one for a single, a walk, a steal, or being hit by a pitch, two for a double, three for a triple and four for a home run. Then add all the bases for the year. Call them aggregates bases.

Conversely, count each out that the player makes -- whether he makes it at bat or is thrown out stealing. If he hits into a double play, charge him with an extra out. (Sacrifices of either kind are ignored since, implicitly, they are a tactical trade of a base for an out).

Compute the total average just like a batting average: divide a player's aggregate bases by his plate appearances (at bats, walks and hit-by-pitches) plus his steal attempts and number of times grounding into double plays.

Ultimately, a statistic -- even one that sounds this reasonable -- is only valuable to us if it reinforces what might be called our own well-informed common sense.

Does the thing give results that seem sensible, and which mirror our perceptions of players more accurately than other stats? Does it support judgments that we previously felt instinctively, but couldn't back up with any other individual stat?

The simplest appeal is to look at the total average all-star teams. If they aren't right, then the stat stinks.

Once again this year, TA gives line-ups of offensive all-stars that are more applealing than would be generated by appealing to any other stat.

For instance, TA gives an AL all-star team (see chart) of Porter, Cooper, Grich, Smalley, Bret, Rice, Lynn, Lezcano, and Baylor. Eight of the nine had 100 RBI, and the ninth (Smalley) had 95.

All nine, in fact, led their position in RBI per at bat.

By contrast, an AL all-star team built on batting average, rather than total average, would include Brian Downing, Bruce Bochte, Paul Molitor, Rick Burleson, Al Oliver and Steve Kemp, who, as a group, drove in 176 fewer runs than their TA counterparts.

In the NL, an all-star team built on batting average would include Bob Boone, Keith Hernandez, Phil Garner, Ray Knight, Gary Matthews, and Dave Parker, instead of TA's Ted Simmons, Willie Stargell, Dave Lopes, Mike Schmidt, George Foster, and Dave Winfield. The total average guys had 102 more homers.

It is TA's strongest claim that it highlights great seasons that other measuring sticks have a hard time finding.

The '79 total average champion was Fred Lynn. How could it have been anyone else? He led the American League in hitting and paced both leagues in slugging percentage and had the majors' second-best on-base percentage. He hit for average, slugged, got on base and produced more runs per at bat than any regular in baseball (edging Don Baylor).

Lynn's only weaknesses were his customary fragility (15 missed games) and only two steals. That's why five players finished ahead of him in aggregate bases -- Schmidt, Rice, Baylor, Winfield, Brett.

If total average measures production relative to opportunity, the aggregate bases give extra emphasis to durability and variety of skills.

Total average is best suited to evaluating those players who bat in the "heart" of the order --third, fourth, fifth and sixth. After all, isn't that, almost by definition, where the best offensive players go?

Men who bat first and second, lacking power, perhaps, but having great speed, do well in TA but better in aggregate bases.

Omar Moreno, Ruppert Jones, Pete Rose, Ron LeFlore, Templeton, Willie Wilson and Lopes fall into this top-of-the-order category.

Some platoon players immediately leap to the total average eye. Oscar Gamble (.657), Champ Summers of Detroit (.657) and Jim Spencer (.613) all batted between 270 and 300 times last year, yet were spectacular. All three are lefty swingers playing in parks with short right-field porches. They were used at home against righties. And they delivered.

A hidden key to Baltimore's pennant was a trio of lesser-known fellows -- Gary Roenicke, John Lowenstein and Pat Kelly -- who had a combined total average of .572 in their 726 at bats. Their TA was just a shade behind the likes of Reggie Jackson and Dave Parker.

Baseball also has an equally unknown subculture of players whose true offensive incompetence is hidden behind respectable batting averages.

Who, for instance, could have been worse in '79 than Baltimore's Mark Belanger, who hit .167 and had a TA of .329?

Oh, lots of folks. The Cleveland keystone combo of Kuiper and Tom Veryzer had the two worst TA's of any regulars in baseball, .305 and .306. At least Kuiper won a gold glove. He'd better.

Mario Mendoza of Seattle, who somehow was allowed to bat 373 times, was even worse with a TA of .266.

Players about whom this author wishes never to hear another word of offensive praise include poster-boy Bucky Dent (.352), millionaire Rennie Stennett (.376), hot prospect Danny Ainge (.311) and a pair of part-time favorites, Roland (Vacant) Office (.397) and Larvell (Blankty) Blanks. (.265).

A special award, however, should go to Willie Montanez. In 310 at bats for the New York Mets -- a team from which he desperately wanted to be traded -- Montanez's TA was .345. Once released from bondage and shipped to Texas, Montanez' total average over his last 144 at bats was .541.

Instead of a hot dog, Montanez's new symbol should be a gold brick.