Much talk has occurred about the Indians starting rotation being a team weakness, because the overall stats aren't that pretty. Each of the Indians five starters (for the sake of this article, Masterson, Kluber, Jimenez, Kazmir and McAllister) has an ERA above 3.50 and all but Masterson have a WHIP over 1.20. Using WAR, the Indians starters rank 16th, 54th, 75th, 97th and 114th among American League pitchers and yet Cleveland is just 3.5 games away from a playoff spot with two months left in the season.
Statistics don't tell the whole story here however, as these undervalued starters have given the Indians a chance to win every single time out. Here is a chart that shows how the Indians starters have pitched (through ERA) during good games (at least five IP and less than four earned runs allowed) and bad games (everything else) and the team's results from those games:
*For Corey Kluber, his first relief appearance was not included, but his second, where he appeared in long relief and earned a win was. Also, his rain shortened appearance against Tampa was considered a good start despite just throwing two innings.
What is being shown isn't really that groundbreaking. Obviously, pitchers numbers will look better during games they play well than in games they play poorly. What should be noted is there are vastly more games that have been pitched well than have been pitched poorly. This shows the primary point that despite not having the greatest personal stats, these particular starters are giving the Indians the best chance to win. Overall, their team record in "good" games is 43-19, while it is just 7-17 in poor starts. If the standards are raised to throwing at least six innings to make it a good start, the overall record jumps to 36-6. This shows the overall importance of the starting pitcher going that one extra inning before turning things over to the bullpen.
The alternative to a starter who pitches fantastically in two out of every three starts and terrible in the other, is to have a starter that actually allowed his average every single game. This would actually make things more difficult for a team, like the Indians, that has trouble scoring runs. By using everything they have in one start, rather than pacing themselves, Indians starters are almost guaranteeing wins (86% of the time) when they throw six innings while allowing three or less runs. The Indians still have some chance at winning in the other starts if the offense actually shows up and scores more than four runs, but since the starters are good more than 50% of the time, they are essentially guaranteeing at least a winning record with this strategy.
Of course all this is assuming this is a strategic decision and not just a random trick of statistics that will soon go away.
Zach McAllister has been the steadiest starter to this point, with just one start qualifying as bad.