MLB Pitching Question (Math)

[DrunkenGoon]

What are the chances that at least one of James Shields (CWS), Alex Cobb(Bal) or Tanner Roark (Was) go on to lose at least two more games the rest of the season? For simplicity assume they all get 3 more starts.


I'm not sure how much probability to assign to no decisions for each start.

Help with the math would be appreciated

 

[Biff41]

Not an expert but I would guess you would take the pitchers win loss rate...assign that a value. Then take the win loss rate of the two teams playing. That would be a second value. Then compare the win loss rate and or ERA of the opposing pitcher...that would be the third value. That would have to be done for all their upcoming games...too complicated to deal with unless your a number cruncher.

 

[Miley Cyrus]

If we go strictly off current win-loss record (far from perfect, I know), then

James shields: 6-16 record. Losing pct: .625 prob losing at least 2 losses = .68359375
Alex Cobb 5-15 record. Losing pct: .75 prob losing at least 2 losses =.84375
Tanner Roark: 8-15 record. Losing pct: .652173913 prob losing at least 2 = .72121311739
Probability of at least 1 pitcher losing at least 2 games is
1-(1-.68359375)(1-.84375)(1-.72121311739)
=1-.31640625*.15625*.2787868826
=1-.0137827988=.9862172012

Of course, after getting toward the end, I realized I didn't include the history of no decisions, which would bring that value down a few percentage points probably.