I wanted to take a moment to explain, or link to thorough explanations, the stats I often utilize for to help with understanding them. While I try to keep my observations fairly easy for most to understand with a heavy reliance on simple metrics like OPS, FIP, and WHIP, it dawns on me that many might not fully understand what each stat I utilize actually refers to. With that in mind, here is the numbers I use and often, why I utilize them.
This page may be updated if I adopt a metric.
Use this navigation to help with this page: Batting/Base-Running | Defense | Pitching | WAR
OPS...Simple and easy. You can go the complicated math way or just simply add OBP and SLG. I find that it's a better tool to compare a player's individual seasons rather than compare players.
OPS+...You often see this at Baseball-Reference. It takes a player's OPS and normalizes/adjusts it to take out variables such as the park factors and league factors. What it arrives at is a number on a scale where 100 is average, anything above is above-average, anything below is below-average. For instance, if a player has a 110 OPS+, we can say that his adjusted OPS was 10% higher than league average during that season. Similarly, a 90 OPS+ means that a player's OPS+ was 10% worse than the league average hitter. It turns OPS into a better comparison tool and though I often use other metrics that do the same thing, but with higher specificity, OPS+ continues to be a good tool.
BB% or K%...Also simple. It's found by dividing plate appearances by walk or strikeout depending on the metric. Another number I like for comparing and contrasting an individual player's seasonal numbers.
wOBA...One of my go-to offensive metrics. It stands for Weighted On-Base Average. Because a real explanation goes beyond my hope to provide brief descriptions, visit Fangraphs for the details. To put it into its most simplest terms, wOBA takes batting average and turns it into a stat that helps to describe the hitter's overall offensive ability. One of its best features is that it takes what we already know about OBP and utilizes it to provide us a workable scale. If you know what a good OBP looks like (.350 seems good), you already know what a good wOBA would be. It's more difficult to compute than OPS, but by weighting each kind of hit properly, it's also more descriptive - yet it's fairly easy to understand from a sense of what's good and what's not.
wRC+...If wOBA is a modern replacement to OPS, wRC+ replaces OPS+. This statistic stands for Weighted Runs Created Plus. It took Bill James' old Runs Created Stat, tweaked it because not every hit is worth the same in different parks in different years, and adjusted it to provide a solid comparison tool. Similar to OPS+, wRC+ uses 100 to signify that season's league average. Once again, if a player as a 110 wRC+, we can say that the player created 10% more runs than the league average player. For a far more detailed understanding, visit Fangraphs.
BsR...I don't utilize this much, though that's possibly a result of a lack of real running threats for Atlanta over the years, but BsR stands for Base Running. It's a catch-all statistic that utilizes other advanced base-running metrics to arrive at a number reflective of a player's base-running ability. Basically, the idea is that while base stealing is surely an important part of base-running, it doesn't tell us the whole story. A player can be a poor base stealer, but a good base runner. Conversely, the opposite can occur. A zero (0.0) in BsR is akin to calling the player a league average base runner. Scores higher or below 0 naturally refer to how well or how poorly the player's BsR compares to league average. One reason why BsR is important for me is that it's utilized in Fangraphs' idea of WAR. You can read more about BsR here.
ISO...This metric was originally popularized by Branch Rickey and his team so it goes back a long time. Its value in today's world is its simplicity. ISO stands for isolated power or slugging and like OPS, you can do all the math to arrive at the total or take the short-cut and simply minus batting average from slugging percentage. When people talk about raw power or maybe in-game power, ISO gives us a number. I like to use ISO to compare a player's individual seasons so see if there was a loss or gain of power.
Quality of Contact Stats (Soft%, Med%, and Hard%)...These numbers can be utilized as a more descriptive understanding of contact over the more simple line drive/flyball/groundball rates - though the latter still have value. Baseball Info Solutions uses a mathematical algorithm that seeks to find the quality of the hit. Any number of being told that the player "squared that one up" or "hitting the ball real well" apply here. Hitting the ball hard does not equal a hit, but it does increase the likelihood of one. For more details, read this article from Fangraphs.
Return to top.
UZR...One of the most widely accepted defensive metrics in a world that still isn't so sure about defensive metrics, UZR stands for Ultimate Zone Rating. Like other comprehensive statistics, it utilizes other metrics to arrive at a comparable scale number where 0 is league average. For a long time, we utilized poor defensive metrics like fielding percentage and errors, but in the wake of the sabermetric explosion, numbers like UZR help us to understand better a player's "ultimate" defensive contribution. You can get lost in the details, but here are a few take-aways. UZRs value is at its highest when the sample size is large. Single-season UZRs can become outliers so three-year UZRs typically have more value. While it's a simple catch-all stat, it should be utilized with other metrics when attempting to make a thorough judgement.
DRS...When you watch a player like Andrelton Simmons on the regular, you may have already been introduced to DRS. It stands for Defensive Runs Saved and I like to refer to it as the quantifiable idea that a player "has RBIs in his glove." Like UZR, 0 stands for league average and also like UZR, it looks at total defensive contributions.
Pitch Framing RAA...Both Baseball Prospectus and StatCorner utilize a similar method of calculating this, but because StatCorner is free, I am more likely to use their number. It stands for Pitch Framing Runs Above Average. Again, 0 refers to league average and this number helps to tell us how good a particular catcher is at "stealing" strikes. Pitch framing can be a valuable tool for a team looking to gain an edge. I wrote an article at About.com on July 9, 2015 about this skill.
dWAR...Found at Baseball-Reference, dWAR tends to be a number I look at when I am seeking a bigger overall impression of a player's defense. Basically, the number uses both DRS and Total Zone Rating, a stat I don't utilize much, to arrive at an overall score relative to 0 of a defender's ability.
Return to top.
FIP...Commonly used in conjugation with ERA, FIP stands for Fielding Independent Pitching. What makes FIP valuable is that judging it does not exist in a vacuum. It looks just like ERA so a FIP of 2.80 is, just like with ERA, a good number. FIP focuses on the Three True Outcomes, or things that the pitcher has more direct control over - HR, BB, and strikeouts (often, HBP is grouped in). The goal is to take out the fielding element and focus more on the pitcher and their actual performance. With that said, FIP has more value as a future indicator than a present evaluation tool. Because, for most pitchers, FIP and ERA grow closer as the season progresses, we can utilize FIP as a tool to suggest that a pitcher might be due to get better because his FIP is significantly lower than his ERA or regression to the mean will increase his ERA closer to his FIP. At least three other variations of FIP exist, though I will largely use FIP only. Expected FIP, or xFIP, adds in factors such as normalized homerun rate and flyball rate to strip away the randomness of pitcher performance. Expected FIP Adjusted, or xFIP-, takes xFIP and adjusts it for park and league factors. Finally, there is cFIP, or Context FIP. It brings a lot of factors back into the conversation such as the batter and umpire and is normalized into a scale similar to wRC+. Instead of an ERA-resembling number like FIP, cFIP normalizes it so that 100 is average, but like a lower ERA works in the same way as a higher batting average, under 100 is better for cFIP. So, a cFIP of 80 is 20% better than the league average cFIP. Honestly, cFIP is the best option of the group, but it often takes an explanation at this point because it's simply not well known.
SIERA...Though it sounds like a bad stripper name, it actually stands for Skill-Interactive ERA. While FIP and xFIP seek to strip away the balls that are put in play, SIERA brings them back in to help better display successful pitchers. It works like FIP with a number that resembles ERA. SIERA is very difficult to understand or even explain, so please read more if you're interested. Like many pitching metrics, it works best as part of a thorough look into a player - less as the be-all, end-all number.
BABIP...While BABIP affects both pitchers and hitters, I tend to focus more on it for pitchers - though it was fun to write about the "King of BABIP" Chris Johnson. BABIP stands for Batting Average on Balls in Play. It's pretty straight forward, though how we use it is important. For me, I use it as a predictor for both hitters and pitchers, though again with a focus on pitching. BABIP comparisons between players are not all that valuable. Instead, I use BABIP to compare current performance to previous performance. If a pitcher has a career BABIP of .310, but it spikes to .370 this year, it might be a sign of something. I stress might because it might only the pitcher has rotten luck. I use BABIP as a jumping off point.
LOB%...Like BABIP, Left-On-Base percentage can be used as a sign of possible future performance. If a player has a high LOB%, especially compared to previous performance, we might think that he's "lucky" and prone to regression. Similarly, if his rate is lower than normal, we might think that he's "unlucky." Again, "might" is the important word because may be a cause for rate fluctuations.
HR/FB...A third potential indicator of future performance much like BABIP and LOB%, the percentage of flyballs that are homeruns might tell us something about performance. Significant variations between the current season number and previous year's might prompt us to search for reasons why.
K% and B% vs. K/9 and BB/9...For the longest time, I used strikeout and walk rates with the consideration of how many strikeouts or walks per nine innings a pitcher had. I've come around to realize that the more accurate representation is found in percentages of plate appearances that end in either a walk or strikeout. While both tend to tell you the same thing, it becomes far easier to spot a significant increase/decrease when we look at the percentages.
Return to top.
Because we can't have things easy, WAR, or Wins Above Replacement, has two different widely-used interpretations. There is fWAR, or Fangraphs interpretation of WAR. There is also bWAR or rWAR which refers to Baseball-Reference's interpretation. Like Mac and PC, both have their proponents.
WAR is a catch-all statistic that is great for comparison both among current players and between generations. It's become more and more accepted as a valuable statistic. The great thing about WAR is that, at its core, it attempts to value a player's overall contributions to a team rather than focus solely on his hitting or pitching. Its usage of defense and base-running makes WAR a number of importance.
Which one will I use more often?
Even though I link to B-R, I utilize fWAR more. My basic reasoning is that I use FIP and UZR and fWAR uses each in their formula. It's important to note that rWAR is often lower than fWAR so the two numbers don't necessarily compare easily with one another. For instance, a player with a 5 rWAR should be considered to have a better season than a player with fWAR. Conversely, while fWAR is the one I use, rWAR divides hits total into oWAR and dWAR (offense and defense) in an easy to find manner, which makes it a bit more convenient.
Neither WAR is better than the other. They are simply different. The user may choose, like I have, to have a preference for one over the other.
Return to top.