I’ve gotten requests since I started writing Mound Visit to make an analytics guide for fans who are intrigued by the new stats but don’t really know where to start. Last week, I wrote part one, examining stats that mostly relate to pitchers. If you missed it, you can find it here.
For part two, I’ll be focusing on stats that are more hitter-centric. Part one is going to be more elaborate, partially because it’s not worth covering WAR and pitch values again, but also because a lot of offensive stats are more straightforward. Unlike pitching, where there are a ton of different stats or peripherals you can latch on to for an individual, hitting can be boiled down to a couple stats to tell most of the story. Obviously there’s more to hitting than just someone’s exit velocity, but a player’s offensive output is usually easier to quantify.
“PLUS” STATS
One of the first analytical stats that reached the mainstream 20-something years ago was on-base plus slugging, or OPS. It offered a way to evaluate a hitter’s input beyond just batting average, factoring in walks and extra-base hits. However, not all OPS marks are created equal.
Theoretically, a player could have a .300 OBP and .500 slugging percentage at Coors Field or a .400 OBP and .400 slugging percentage at Petco Park. Both of those players had an .800 OPS, but the second player had the better offensive season because he played in a tougher hitting environment and 100 points of OBP is worth far more than 100 points of slugging. (Rule of thumb is one point of OBP is worth about three points worth of slugging.)
That’s where OPS+ and wRC+ (weighted runs created plus) come into play. Baseball Reference uses OPS+ and FanGraphs uses wRC+. Both stats factor in the player’s OBP, slugging percentage and home ballpark and compares it to the league’s offensive output. Both are put on a 100 scale where 100 is league average. If a player has a 120 OPS+, that means they produced 20% more offense than the league average hitter. If they have a 70 wRC+, that means they produced 30% less than that average hitter.
OPS+ and WRC+ are rate stats – like batting averages – so you do have to be mindful of sample sizes when comparing players. But assuming a player has had enough at-bats to create that decent work sample, then here’s how you can evaluate their wRC+ or OPS+:
Awful: 60 (Kevin Newman, 56 OPS+)
Poor: 80 (Ke’Bryan Hayes, 87 OPS+)
Average: 100 (Colin Moran, 98 wRC+)
Good: 120 (Adam Fraizer, 128 wRC+)
Excellent: 150 (Bryan Reynolds, 146 OPS+)
The Pirates had a team wRC+ of 83 last year, 29th in baseball. Given that the Pirates finished at or near the bottom of every offensive category, that wRC+ is justified.
I tend to use OPS+ more in stories because I think the terminology is a little more familiar. The differences between the two are minimal, so it’s just a matter of preference, usually of if you prefer Baseball Reference's or FanGraphs’ version of WAR. Just about every player’s OPS+ and wRC+ numbers will be within two or three points, which is minimal.
And to tack one last thought onto the pitcher stats from last week, Baseball Reference also has ERA+, which functions the same way with a pitcher's ERA.
BATTED BALL DATA
We touched on ground ball and fly ball rates last week and how to find them on FanGraphs. It goes without saying that fly balls and line drives are better than ground balls, so we’ll leave that be. We also discussed exit velocity last week and how more is almost always better.
That’s not the case for launch angle. You want to catch the sweet spot of the bat and drive it. Launch angle measures how a ball leaves the bat, ranging on a 180 degree radius of 90 to -90 degrees. A ball with a launch of angle in the negatives is beaten into the ground. If it’s too high, it’s a pop up.
Keep that in mind because average launch angle is not as reliable a stat as average exit velocity. In the latter, every quality batted ball is going to be reflected in a higher average exit velocity. For launch angle, you could potentially want a higher potential angle, meaning less ground balls, but a lower average angle could be good too because it means they hit fewer pop ups (see: Mike Trout).
Rather than rattle off a couple Pirates average launch angles, let’s take a look at what a good launch is. This doesn’t take exit velocity into account, but these are the slugging percentages based on launch range:
Baseball Savant
Most damage is done in that 20-35 degree range. Assuming the exit velocity is high enough, batted balls in that range are referred to as “barrels.” Baseball Savant keeps track of what percentage of at-bats end in a barrel, which is the ideal batted ball.
DEFENSE
Out of all aspects of the game, defense remains the toughest to quantify. For years errors and fielding percentage were deemed good enough metrics to use, but it doesn’t take into account a fielder’s range. Why should someone be penalized with an error when they had the ability to get to a ball deep in the hole but couldn't complete the play when a less gifted defender would just watch it go by for a hit?
A good, albeit sometimes flawed, stat to use is defensive runs saved (DRS). The name is self explanatory. The calculations aren’t. Here’s how DRS is calculated according to Fielding Bible. Sports Info Solutions is the source of FanGraphs’ DRS leaderboard, and their formula was tweaked in 2020 to account for a player’s starting position, taking away runs that were originally deemed as saved because of a shift rather than a player’s defensive ability.
There can still be some outliers, like how Reynolds had -5 defensive runs saved in center last year despite being graded well by Baseball Savant and being nominated for a Gold Glove. Defense is tricky to calculate.
That is especially true for catchers. FanGraphs tracks the number of runs a catcher saves through framing (measured to a tenth of a run, since it’s impossible to know for sure how every at-bat would have played out if that call on the edge went the other way). Baseball Savant tracks catcher framing too, and actually breaks it down by different parts in the zone. Baseball Prospectus may not be the analytical giant that it was 10 or 15 years ago, but their Fielding Runs Above Average (FRAA) for catchers may be the best way to evaluate the position because of how they weigh blocking runs.
Defense goes into WAR calculations, with FanGraphs using Ultimate Zone Rating (UZR) and Baseball Reference using Sports Info Solutions’ DRS (dating back to 2003. Before then they used Total Zone Rating). Of the two, the latter is better in my opinion because UZR is more of a predictive tool to evaluate what they will do defensively in the future rather than assessing how the player actually performed in the field. I prefer WAR to be more concrete in its calculations.
DRS is an easy and solid tool if you know where to look, but Baseball Savant is where you can really geek out on defensive stats. You can check to see what depth a player averaged last year (did you know Wilmer Difo played nine feet shallower in center field last year than Reynolds?). What defensive shifts did a hitter face? What catch probabilities did an outfielder have? What jump did they get on a fly ball?
The one stat from Baseball Savant that I think carries the most weight is outs above average (OAA). It’s their version of DRS, but it’s all based off of hit probabilities. It rewards players who make the routine plays, while also incentivizing highlight reel catches.
For example, if a batted ball to second has a 95% chance of becoming an out and is converted into being an out, the fielder is credited for that 5% boost in defensive value. If they botch it, they’re penalized 95%. Meanwhile, if they convert a five-star play that only had a 10% chance of being an out, that’s a huge boost for their defensive value. If they can’t get it, it’s not heavily weighted since the odds were not good.
Baseball Savant keeps track of OAA for every position besides pitcher and catcher. Going back to Reynolds, his 9 OAA were 10th-best among outfielders last year. Again, defense is hard to calculate. OAA rewarded the outs he made more while DRS penalized the plays he couldn’t make more.
To rattle off some other players’ OAA and DRS totals from last year: Hayes had 16 DRS and 13 OAA, both of which were tops among National League third basemen. Yoshi Tsutsugo had -5 DRS saved in the outfield with the Pirates in just 150 ⅓ innings. (To look at it another way, he cost his team an extra run in every three games he played in the outfield. That’s why he’s considered a first baseman now.) Ben Gamel struggled making plays to his left and had -10 OAA in left field last season. That can be a problem with PNC Park’s left field notch.
Each team runs its own defensive models for their own internal evaluation, and from what I’ve gathered, OAA is the closest to what some teams use. Like OPS+/wRC+ and fWAR/rWAR, the difference between OAA and DRS isn’t too much usually. Either is a good tool, as long as you know how to apply it. That’s the key for using any advanced stat.