In 2007, Carlos Lee hit 32 homeruns, knocked in 119 RBIs, made the All-Star team and finished in the top 20 in MVP voting. And Mark Ellis? Well, he didn't do any of those things. But he's a much better baseball player than Carlos Lee. Seriously.

Let's start with offense, where Lee created more runs than Ellis, 111 to 79. That includes contributions from singles, home runs, walks and stolen bases. But Ellis did play in a worse hitters' park and hit against better pitchers (the AL is a stronger league than the NL). To account for that, I'll hold Lee against a higher replacement level. The result is that Lee's offense was worth 41 runs above replacement level as opposed to Ellis' 19 runs. (If you want to see the math drudgery, I'll be happy to post it in the comments section.)

On defense, Ellis plays the more demanding position. Second base is about 5 runs more difficult than left field over a full season. Just think about how ugly Lee would look at second base. On top of that, Lee plays his position poorly while Ellis is a stud. How bad is Lee?

He was about 15 runs worse than average in 2007, even worse than his track record of -7 runs per season in both 2005 and 2006. Ellis, on the other hand, was 25 runs better than average last year and has averaged +15 runs per season over the past four years. In order not to give Ellis too much of an advantage based on just 2007 numbers, let's put Lee at -10 runs and Ellis at +15 runs. That's a 25 run advantage for the Athletics' second baseman, and 30 once you add on his position adjustment.

Time to bring it all together. Lee started out 32 runs better than Ellis when looking at raw runs created on offense. That gap closed to 22 runs when adjusting for the difficulty of hitting in the AL, hitting at the Oakland Coliseum, and the fact that Ellis created his runs while using up fewer outs. When fielding is considered, Lee's lead turns into an eight run advantage for Mark Ellis. Overall, Lee is worth about 26 more runs than a replacement player, which is slightly above average. Ellis sits at about 34 runs better than replacement level, bordering on All-Star selection.

So what's the moral of the story? Simply put, there's more to winning baseball games than raw offensive numbers. For one, the ballpark and league contexts of those numbers are important. Defensive position is also important - if Lee could play second base, he'd help his team by allowing some other all-hit, no-field freak to take up space in left field. And lastly, fielding is more than a footnote - one extra play made per week adds up to more than 15 runs per season. Sure, the range of hitting talent is a lot wider than fielding talent, but isn't the old cliché "defense wins championships"?

The mistake of overrating Carlos Lee types and ignoring Mark Ellis types get compounded when teams throw huge money in the wrong direction. Lee will average a $15 million payday over the next five years even though he's a $10 million player at best. What could the Astros do with an extra $5 million per season? They could start by paying Mark Ellis' salary.

By skyking162 of www.DugoutCentral.com