DFA, Mikev, PaulThomas and Danmerqury I apologize. You guys are better than I am at this stuff but I am tired of moderating comments that are flame wars between the internet versions of scouts and saber fans. Hopefully you guys will release something better soon. I am only scratching the surface.
A brief commentary first- I have the 'ol Master of Business Administration degree rotting in a box somewhere. It isn't from a good school. It's from National University. I tell you this because I don't want anyone thinking i am trying to be a braggart about my superior intellect. But one of the key lessons I learned in completing this degree program was that statistical analysis and human observation are both key to the success of any business. So.... here is a list of stats I like to look at, the plain English of what they tell me and then the formula (if I can find it) that is used to create them.
The two I always find myself reviewing (position player):
wOBA- Weighted On Base Average. This is like On Base Percentage and Slugging only it takes into account all possible outcomes (walks, singles, doubles, etc.) and each are weighted so that it is a more accurate representation of a player's offensive ability and expressed in a form similar to OBP (because we are all familiar with OBP). This was created by Tom Tango, who rocks. Here is the formula:
((0.72 x NIBB) + (0.75 x HBP) + (0.90 x 1B) + (0.92 x RBOE) + (1.24 x 2B) + (1.56 x 3B) + (1.95 x HR) / PA
NIBB= Non Intentional Base on Balls, RBOE= Reached Base On Error and the rest are standard back of the baseball card stats.
UZR/150- Ultimate Zone Rating per 150 Games Played. You can think of this as an entirely more relevant version of fielding percentage, or something. UZR is complicated. The easiest way to explain it is to think of the baseball field carved up into a huge pie chart. Then, each slice of pie is assigned an average number of balls that are hit to that "zone." Next, you look at how many outs are recorded by a given player when playing a particular position. Last you measure that against the expected number of outs based on the zones covered. If they make more outs than expected the UZR is a positive number. If they make less, it is a negative. The per 150 games is just a method of normalizing the data. I believe this is an imperfect measure. It is clearly not as good at evaluating defense as wOBA is at evaluating offense. But it seems to be the best available metric.
I don't know the formula for this one. Even if I did, access to the data required to calculate it is pretty hard to manage on our own. Just go to fangraphs and look at it :)
1 For Pitchers:
FIP- Fielding Independent Pitching. This thing is AWESOME. Basically think of it as ERA only more objective and relevant. Like wOBA above, this is a pitching statistic that measures the things a pitcher has total control over and weights them appropriately to give a better representation of overall performance. This is another Tom Tango creation (or at least the formula I will provide is). The formula is:
I would be remiss if I didn't mention Voros McCraken's DIPS (Defense Independent Pitching Statistic). It is sort of the OG of awesome pitching stats that make ERA and Wins irrelevant.
1 for anybody:
WAR- Wins Above Replacement. Replacement Level stats confuse the hell out of most people, but the general idea is to place a value on a real player's performance as measured against a fictitious replacement. The fictituous replacement level player is assumed to be a Triple A call up, below average major league player. So a team full of replacement level players would suck in a huge, huge way. WAR is a way to put a value on the total number of runs a player contributes to a team with every 10 runs resulting in a theoretical "win." One last important thing to note is that there is also an adjustment made to the numbers based on the position played by the player being evaluated. The takeaway? This stat evaluates a player's total contribution to the team taking into account offense, defense and the position the player played.
To cite examples... Jack Cust has an excellent wOBA, but a crappy UZR/150. Conversely, Ryan Sweeney has a very good UZR/150 in RF and a less impressive wOBA. Both of these would be taken into account to assign the runs they provide to the team and consequently both would factor into their WAR. So I think of WAR as the ultimate summary of a player's performance.
These 4 stats are just the surface. They are the summaries I use to understand a player's performance. There are caveats... for instance a player who has a WAR elevated by UZR alone should be viewed with some measure of skepticism, because UZR is less concise than wOBA. But overall, these stats will provide you with a good understanding of player's value relative to another player. Simply put, we want more WAR.
Let the constructive conversation begin!