45/Dan - If you want old baseball prospectuses, I might be able to help you - up to and including the rsb postings that essentially comprises what prospectus looked like the year before it became a book.

I've got to say that linear weights blows a lot of the stuff James was doing out of the water.

I also was fascinated when Linear Weights first appeared with Hidden Game back in 1984. Not only could you use LW to compare/rank players, but for the first time I felt you could see actual numbers backing up certain old adages in the game, such as "good field, no hit". Seems like every team had one of these guys - in some cases, it looked like the defense really was good enough to offset a weak bat (Ozzie Smith) but, in a lot of cases, teams were killing themselves putting players on the field that just cost way too many runs to have in the lineup.

Now we are in the WAR era, almost 30 years later. B-Ref has essentially adopted the offensive component of Linear Weights in the Batting Runs calculations and most people seem to think that is reasonably accurate. The defensive side of WAR, however, leaves a lot to be desired, even with using play-by-play data, and a lot of people are so disenchanted with dWAR that they abandon WAR altogether for being too inconsistent. For those who really like to dig in the numbers I highly recommend Michael Humphreys' book "Wizardry" - he goes into great detail about how he computes DRA for evaluating players defensively. At least with his version you can see all the formulas for how the numbers are computed. You may not agree with his conclusions but at least you can see where they came from.

One thing I thought was kind of funny last fall when the MVP debates were happening - according to B-ref WAR, one of the reasons Trout trumped Cabrera so thoroughly was that not only was Trout the much better baserunner and fielder - no one was really arguing that point - but that Trout was the better hitter. According to Batting Runs, Trout had 54 Rbat and Cabrera only 52. This really set a lot of the Cabrera fans on fire - how could a guy who won the Triple Crown not be the best hitter in the league? I think it's primarily the Park Factor adjustments that leapfrogged Trout over Cabrera, and PF adjustments have been part of offensive LW for decades now, so that's really nothing new. But it sure set off a lot of the Miggy supporters, who have now decided that not only is dWAR unreliable, but oWAR as well.

So maybe we're back to the drawing board trying to find the best measurement stat. Maybe WAR isn't the answer, but I think we're at least heading in a better direction than we were 40 years ago.

For those who really like to dig in the numbers I highly recommend Michael Humphreys' book "Wizardry" - he goes into great detail about how he computes DRA for evaluating players defensively. At least with his version you can see all the formulas for how the numbers are computed. You may not agree with his conclusions but at least you can see where they came from

Seconded. Given that, without a time machine, nobody can reconstruct zone data for games that were never observed for it, something like DRA is the only recourse to compare defense across eras, and Humphreys's book is a wonderful exercise in the logic behind doing so.

how could a guy who won the Triple Crown not be the best hitter in the league? I think it's primarily the Park Factor adjustments that leapfrogged Trout over Cabrera

Although it's also true that Trout led the majors in Runs Scored despite missing 23 games of the season. To put that forward as an MVP credential is so old-school, it's new-school :)

how could a guy who won the Triple Crown not be the best hitter in the league?

In 1992, Barry Bonds hit .311/34/103. He had a 204 OPS+ and 8.2 oWAR. Gary Sheffield didn't win the TC, but he was in the hunt all year. He finished with a .330/33/100 slash line, winning the batting title, and missing the TC by 2 HR and 9 RBI. He had a 168 OPS+ and a 6.8 oWAR. 2 more HR and 9 more RBI would not have gotten him close to Bonds.

In 1992, Barry Bonds hit .311/34/103. He had a 204 OPS+ and 8.2 oWAR. Gary Sheffield didn't win the TC, but he was in the hunt all year. He finished with a .330/33/100 slash line, winning the batting title, and missing the TC by 2 HR and 9 RBI. He had a 168 OPS+ and a 6.8 oWAR. 2 more HR and 9 more RBI would not have gotten him close to Bonds.

True, but Miggy also won the slugging percentage crown and the raw ops crown that all the stat guys are talking about (and was only behind trout by .006 in obp ) I don't think the Cabrera supporters are as old school as people seem to portray them as.

1.4 WAR? What is that, @ 14 runs? When you're hitting that many home runs, how accurate are the runs estimates?

1.4 WAR? What is that, @ 14 runs? When you're hitting that many home runs, how accurate are the runs estimates?

Well, it's actually more than that. I mistakenly used oWAR, which includes a defensive position adjustment. Sheffield, who played 3rd that year, got a +1 for position, while Bonds who played left, got a -6.