The Walter Johnson example in post #2 only covers his 1910-19 seasons, so I did a full career chart for him first:
Walter Johnson: 417-279, .614
Others: 1142-1330, .462
Total: 1559-1609, .492
122-point winning% gap between himself and the teams he pitched for, and Johnson accounted for 26.7% of his teams’ wins.
Then I did one for Ted Lyons, throwing out his first and last seasons since he didn’t pitch much in those. The results:
Ted Lyons: 257-225, .533
Others: 1073-1339, .449
Total: 1330-1564, .460
73-point winning% gap between Lyons and the teams he pitched for, and he accounted for 19.32% of his teams’ wins… but take that latter statistic with a grain of salt because he played in a later era when starters pitched a lower percentage of complete games.
Here’s Pete Alexander, not counting his last season when he didn’t pitch much. 1926 was a split season between the Cubs and Cardinals and the math gets hairy there, but I used the Cubs’ record up to June 21 and then the Cardinal’s record from June 22 when they acquired him.
Pete Alexander: 373-205, .645
Others: 1168-1121, .510
Total: 1541-1326, .537
108-point winning% gap between himself and the teams he pitched for, and he accounted for 24.2% of his teams’ wins.
Here’s Seaver:
Tom Seaver: 311-205, .603
Others: 1279-1430, .472
Total: 1590-1635 .493
110-point winning% gap and he accounted for 19.6% of his teams’ wins. It’s been rarer to see that second stat up high in the last 50 years since the five-man rotation has come in.
Carlton was still on the Phillies when they became really good later on, so I knew his career numbers wouldn’t look anything like 1972 for this question. I calculated them anyway though. His first and last seasons weren’t counted.
Steve Carlton: 329-243, .575
Others: 1493-1419, .513
Total: 1822-1662, .523
52-point gap, accounted for 18.1% of his teams’ wins.
And a likely future HoF’er that nobody’s named yet, not counting his first year:
**Roy Halladay: **202-105, .658
Others: 1040-1082, .490
Total: 1242-1187, .511
147-point winning% gap and he accounted for 16.26% of his teams’ wins. Looking over the numbers of those 2000s Jays teams, the hitting usually wasn’t too bad but they were thin on pitching early in the decade and had some bad Pythagorean luck as well. When you’re looking at a fairly short career I guess that luck factor can make the numbers kind of swingy.