> According to Lee Sinins sabermetric encyclopedia,
> 1 Mike Schmidt 515
> 2 Eddie Mathews 502
> 3 Graig Nettles 378
> 4 Matt Williams 370
> 5 Gary Gaetti 360
> 6 Ron Santo 337
> 7 Ron Cey 312
> 8 Vinny Castilla 303

In reply to CJ on SABR-L, i explained what can be done by following
the sample queries that are included in the MS Access database
distributed by Sean Lahman. I closed,
>>
But CJ Haddick needs "all 3b with more than 100G" to begin with. That
is not easy, for the database does not identify batting records with
fielding positions as it identifies them with teams. Indeed, beyond
the scope of a particular lineup there is no agreement on what a "3b"
or thirdbaseman is.

It is easy to select from the Fielding table everyone with 100 or more
career games played at thirdbase. . . . But selecting every "3b with
more than 100G" is like inventing a new way to catch fish where you
aren't sure whether a whale or a shark or a shrimp is a fish.
<<

Evidently, Sinins identifies players with fielding positions, at least
at the career level. That can only be controversial. In real life,
you need to think about what you mean --probably in reference to real
players such as Harmon Killebrew, Dick Allen, Tony Perez, Joe Torre,
and someone who played 40% third 30% left 30% right-- *and then* use
some database management ingenuity.

For homeruns hit as a thirdbaseman (where alone I know that people
speak of alltime records) you need game-level detail that is available
only in the SABR Home Run Log. That covers all but a few home runs in
major league history from 1871, with data at the inning level.

Paul Wendt