.271 on an off year is terrible?
Wow.
You do realize Kemp batted .270 and only played in 70 something games last year, right?
Yes, terrible, if you know how to look at hitters.
You are giving me batting average as a measure of a hitter's worth? Did we time-warp to 1950 or something?
Batting average is a worthless statistic. It ignores a large segment of every hitter's at bats. Kemp had 27 plate appearances that aren't measured by batting average last year (in limited time). Andrus had 78 plate appearances that the batting average statistic completely ignores. So how in the world would I justify a statistic that ignores, for example, 78 of a player's plate appearances? That's an extremely significant gap in the measure of a player's performance.
In my opinion, there are two things you can look at to measure a player's performance:
One, getting on base. In other words, not making an out. The only way a defending team can end your inning is by making outs. Therefore, a player's primary function is to not make an out. If you avoid making outs for (at most) 4 batters in a row, you will score, and it doesn't matter
how you avoided making outs (meaning you don't even have to have 1 hit). What that means is that batting average is woefully inadequate. It ignores hit by pitches and more significantly it ignores walks.
Two, hitting for power. In other words, the best at bat a hitter can have is a home run. The second best is a triple. Etc.
The measure for how well hitters do those two things is OPS. It combines how well a hitter gets on base with how well a hitter hits for power. OPS+ converts a players OPS to measure it against how the rest of the league performs (100 OPS+ is league average for a given year, so if you are above 100, you did better than the average major league hitter, if you are below, you did worse).
Therefore, last year, in his "down year," Andrus had an OPS+ of 81. That is significantly below the league average of 100. It is important to note that, despite your claim that Andrus had a "down year," his very, very below league average OPS+ of 81 is only 3 points below his career average of 84. The best Andrus has ever done in his career is 94 OPS+, still 6% below league average. Other than that one season - his high water mark of 94 - Andrus has had an OPS+ of 89 or below.
Sorry, but that's bad. It just is. It means that he is neither proficient at hitting for power nor does he get on base nearly enough to make up for it. He is a bad hitter. His batting average (nothing special to begin with) is an empty statistic, signifying nothing.
Now Kemp. Kemp actually did have a down year last year. Kemp only played in 73 games, so he was hurt, which may explain his drop in production. His OPS+ was 105, still 5% better than league average, but well below his career 126 OPS+. However, Kemp is still 28, presumably in his prime, and his previous 2 seasons his OPS+ were 147 and 172, so you would think he is likely to rebound and have several more great years, assuming health. Even if he doesn't, his 105 OPS+ - his worst season, essentially - is still a good bit better than Andrus'
best season.
Ignore the batting averages. If you look at the measures that matter - how well they get on base (quantity) and how effectively they hit the ball when they hit it (in other words, quality hits), it's no contest. Kemp is a far, far superior hitter. Kemp is well above league average; Andrus is well below.