As the resident stats geek (at least, unless there's another one being kept in the basement somewhere that I'm not being told about), I figured I owed it to the Goat Rider community to share a bit about the specific sorts of stats I use to evaluate a player, before I start slinging around an entire alphabet soup without wild abandon.

The first question you're probably asking is, "Won't you sabermetricians stop trying to ruin baseball?" The answer is: not until the game is played by robots, as was intended by Reason (sabermetricians, after all, don't believe in deities, being souless autonomons).

The next question you're probably asking is, "What's wrong with our current statistics?" And there's quite a few things, actually.

It should be noted that early baseball stats predate such things as the "baseball glove" and the widespread availability of dedicated baseball fields - baseball was played in tall grass by barehanded fielders, with the pitcher throwing off flat ground. It was almost a different game altogether from the modern game of baseball.

Further complicating matters is the fact that baseball statistics treat baseball as though it were a different game altogether - that game being cricket. British journalist Henry Chadwick developed the overwhelming majority of traditional baseball statistics, and he did so mostly by using the statistics that existed at the time for cricket. Cricket was played with two bases instead of four, and didn't have any walks - so it's no surprise that extra-base hits and walks aren't accounted for in batting average.

Looking at the traditional "triple crown" stats presents us with two problems. RBIs tell us how many runs a team scored, but fail to do a good job of seperating the impact of an individual player from that of his teammates. Meanwhile, batting average and home runs tells us little about how many runs a team will score; last season the Rockies lead the NL in batting average and the Brewers lead the NL in home runs, but neither team scored the most runs - the Phillies did.

The Phillies did lead the league in OPS, however - On-Base Percentage plus Slugging. OBP and SLG are two stats that  you're seeing a lot more of in recent years. OBP is simply the number of times a player reaches base (either by a hit or a walk) divided by the number of plate appearances. It's important to note that OBP is per plate appearance, not per at bat - the official definition of an at-bat being fraught with all sorts of absurd technicalities. [Trivia note: in 1887, walks were included in batting average as calculated by the league. So OBP is one of the oldest baseball stats around.] Slugging is simply total bases divided by at-bats.

The great thing about OPS is how closely it correlates with actual team run scoring. Even better is something like OPS+, readily found on sites like Baseball Reference. OPS+ takes a player's performance, and puts it in the context of his league and home park. It's rather obvious that a player that plays half his games in Coors Field is going to put up very different numbers than if he played half his home games at Petco Park; there's no reason to ignore this when evaluating a player's performance.

I should note that while OPS is good, it's not perfect. In Part II, we'll look at where OPS falls short, and look at something that works better.