There are various theories on how a manager should construct his line up card.
A group of statisticians and sabermetric guys have proposed that the numbers show that:
A. Lefty/right platoon situations do work. Right handed hitters hit lefty pitchers better; and vice versa.
B. High on-base percentage players are more valuable than free swingers (i.e. hitters who cannot get on base by walk or hit-by-pitch).
C. Team speed at critical positions is important to avoid double plays (which destroys run production opportunities).
The two critical stats in their minds are
OBP = (hits + walks+ hit by pitches)/plate appearances
OPS = OBP + (total bases/at bats)
There is a theory that this is the format for a good National League line up:
1. Lead off hitter should be the highest OBP hitter on the team.
2. Second hitter should be the second highest OBP hitter, preferable with speed to avoid double plays.
3. The third hitter should be the highest OPS hitter on the team.
4. The clean up hitter should be the second highest OPS hitter on the team. The #3 and #4 hitters are the main gears in driving in runs.
5. The fifth hitter should be the third highest OBP hitter or third highest OPS hitter if statistically close.
6. The sixth hitter should be the third or fourth highest OPS hitter on the team.
7. The seventh batter should be the best OPS starter left.
8. The eighth hitter should be the best OBP starter left, so that the pitcher in the #9 slot can sacrifice him into scoring position to have the best hitters on the team come up at the top of the order.
On the other hand, shouldn't the best line up card have the best hitters at the top of the order?
If the key stat is OPS (which incorporates OBP into the formula),
the true meaning of OPS can be condensed as follows:
OPS = H/SF + TB/AB
A random player is chosen to discuss: Starlin Castro.
In 2013, he played in 161 games. He had a .245 BA, .284 OBP and .631 OPS. He hit 10 HR, 44 RBI and 9 SB with a negative 0.6 WAR.
Castro's 2013 condensed OPS = 163 H/1SF + 231 TB/666AB
Castro's condensed OPS = 163 77/222.
What jumps out is that this result nearly matches Castro's HIT total.
In 2013, Anthony Rizzo played in 160 games. By comparison, he hit .233 BA, .323 OBP, .724 OPS with 23 HR, 80 RBI, 6 SB and 2.6 WAR. By all measures, Rizzo had a better season than Castro.
Rizzo's 2013 condensed OPS = 141 H/2SF + 254 TB/606AB
Rizzo's condensed OPS = 70 557/606
What jumps out is that this number is nearly one-half of Rizzo's hit total (which is skewered by the fact he had one more sacrifice fly than Castro did).
If you believe that a sacrifice is a "productive out," then it should not have so much negative drag in judging a player's offensive production.
Statistics can be manipulated to achieve certain results. Baseball is another example of trying to glean a bigger pitcher from smaller quantitative analysis of putting game action into numbers.
But one cannot lose simple common sense. In order to have the best line up, the key statistic for all hitters is HITS. The team's best hitters should be at the top of the order.
Based on last year's stats, the 2013 Cubs order should have been:
1. Castro, ss 163
2. Rizzo, 1b, 141
3. Schierholtz, rf, 116
4. Castillo, c, 104
5. Barney, 2b, 104
6. Soriano, lf, 92
7. Valbuena 3b 72
8. DeJesus, cf, 71
