I know it really isn't over yet, but the Twins have a 6 game lead. On July 20, the White Sox had a 3.5 game lead.
The Twins are 39-16 since the All-Star Break. If they had played .600 since then, the two teams would be tied. They were 46-42 prior to the All-Star break. Who could have predicted that they would play .700 since then, especially without Morneau? The Sox only went 30-26. That should have been enough to keep them in contention.
The Twins have allowed an OPS of .682 since then while the league average is .727. The Twins have allowed a .312 OBP and a .371 SLG while the average has been .323 & .404. Before the All-Star Break, they allowed an OPS of .745
The Twins batters saw their OPS jump from .762 pre All-Star to .779. So their OPS differential went from .017 to .097 before and after the break.
They have also beaten the White Sox 10-5 this year. If the White Sox were 7-8 vs. the Twins this year, they would only be two out. If they were to take 2 out of 3 in this upcoming series, they would only be 1 game out and the season series would finish at 9-9. So in a way, the Twins have just taken it away from the Sox by doing so well when they went head-to-head.
For their part, the White Sox hitters have had an OPS of .780 since the All-Star break while their pitchers have allowed a .735 OPS. A .045 differential is not bad. The relationship between OPS differential and winning pct is about
Pct = 1.25*OPSDIFF + .5
That would give the Sox a .556 differential and 31.25 wins. The Twins project out to .621 or 34.17 wins. In reality, the Twins have been 9.5 games better than the Sox in the 2nd half. The OSPDIFF says the teams' W-L record to be:
That makes the Twins 3.42 games better since the All-Star break. The difference could be that the Twins have made 25 fewer errors this year. Also, in the 2nd half, the White Sox OBP differential is only .007 while it is .033 for the Twins. Since OBP is more important than SLG, this could be the key.
In the 2nd half the Sox have outscored their opponents 284-244. The Twins have outscored their opponents 284-196. That gives the Sox a Pythagorean .575 pct or a 32.22-23.78 record. The Twins project out to a .677 pct. or a 37.26-17.74 record. That predicts that the Twins would have been about 5.5 games better than the Sox since the break.
This implies that the Twins might be winning some close games that the Sox are not. In fact, since the All-Star break, the Twins have a 4-1 record vs. the Sox in 1-run games. A quick "eye balling" of Baseball Reference shows the Sox to be 7-11 since the break in 1-run games while the Twins are 15-7. That is 6 games better, the lead the Twins have.