I have literally done thousands of platoons on SOMO over the years in all every SOM game that they have put out. To not use platooms is to give away one of the few ways you have to legally circumvent the salary cap, along with park effects, etc. I usually would defer to Adam Katz and I think he is on the right track, but where I would disagree is that:
1)I would not say that HAL rarely pinch hits, even in the ninth inning. I would say pinch hitting before the ninth inning is rare, but I think it's pretty common in the ninth inning. Not sure about the eighth inning.
2)I would say the jump you get in offense definately outweighs any perceived disadvantage. The people who are expecting their platoons to face 100% of their preferred side are living in a fantasy world(pun intended). That being said, I pretty commonly get guys facing up to 90% to their preferred side unless they are forced to play do to injury. What more can you realistically ask for than a guy facing around 90%(usually vs RHP) to his preferred side? And, if you get a platoon to face that high a percentage, I do not see how the jump in offense could be outweighed by any problems with the platoon playing occasionally against their bad side unless you created a bad platoon, then shame on you.
Here are some examples just from my current teams:
1999(just won championship last night)http://onlinegames.strat-o-matic.com/team/803671Two successful platoons:
Helton-90% vs rhp's/Garcia 67% vs lhp's 120R's/150RBI Cost = $7.31
Hundley-92% vs rhp's/Hernandez 53% vs lhp's 97R's/123RBI Cost = $2.29
I realize the 1999 game has trumped up offensive stats, but the splits are valid anywhere. Garcia would have a better split but he also backed up 3B and took a few turns in the OF and at DH. Hernandez filled in for Hundley for at least a couple pretty good injuries to slide down to 53% vs lhp. As I said, the vs LH side of the platoon seems to always have the worse split.
I could go on and on posting examples from my current teams for different games, but I am confident they would basically show the same results. Both guys above having 90% splits is no fluke and how can that NOT be benefit you towards winning and being successful?