You *can* buy the book with the breakdown, if you like. 1988 Baseball Abstract, pages 45-47. I bet you could find a used copy for a couple bucks.
W% top 2 pitchers: .593
W% rest of team: .491
Percentage of team wins from top 2: 38%
Historical difference between top 2 and rest of the staff: .106 (87 Twins were better than average!)
Historical percentage of wins from top 2: 41% (Twins had more wins from the non 1-2 than average!)
The 1908 Cubs are often considered the best team in MLB history. The difference between their top 2 and rest of the staff: .227. The difference between .102 and .227 is substantial!
James doesn't stop there, he notes that every team has a weakness and he identifies the Twins weakness as missing a #3 pitcher. He does this by throwing away all wins by 1-2 pitchers for championship teams and finding the Twins were under .500 with the rest of the starters. Although not unheard of (and the Twins were far from the worst), the difference maker then became the lack of a #3.
James concludes that the 1-2 v. the rest of the staff was "absolutely normal" for a championship team. He goes on to state that the losing team, the 87 Cards, had 5 great starters but comparatively bad 1-2 starters! The key to a championship is 2 dominant starters.
I had that book at one time - it didn't make the cut on one of my moves.
Les Straker was the #3 starter, and in 154 innings his ERA+ was 106, which is above average.His W-L was only 8-10, and his SO/9 was an unsustainable 4.4, but I find it hard to believe he was the problem when you compare him to the disasters that attempted to fill the 4th and 5th starter slots.
Any analysis based on W% of starters is flawed, and that was in the early days of baseball analytics. Straker was right in the middle of two groups of extremes, and if you lump him in with either group you will have a Sesame Street "one of these things is not like the other" situation."Throwing away all wins by 1-2 pitchers for championship teams", combining the stats of the rest of the starters, and then determining "the difference maker [for the '87 Twins] then became the lack of a #3" just isn't fair to Straker.
I bet if you combined the top three starters for championship teams, the Twins would still be in the normal range for championship teams, but doubt very many championship teams gave 2/5 of their starts to a group of 4 guys who all were well below replacement level. (They combined for -2.9 WAR. Straker's WAR was 2.2)
Edited by gil4, 16 September 2017 - 10:27 AM.