Two hitters both finish a season batting .275. One walks 90 times and slugs .480. The other walks 15 times and slugs .350. Batting average calls them equal. Nobody who actually watches either player would.
The formula is the problem: hits ÷ at-bats counts a walk as nothing (it's not an at-bat), counts a bloop single the same as a bases-clearing double, and says nothing about power or plate discipline at all. It answers one narrow question — how often does a ball in play become a hit — and generations of box scores treated that as the whole story.
The short version: on-base percentage adds walks and hit-by-pitches back in; slugging weights hits by total bases; wOBA and OPS+ go further and weight every outcome by how much it actually contributes to scoring runs, adjusted for park and era. Batting average measures contact. These measure value.
What batting average leaves out
| Stat | What it counts | What it misses |
|---|---|---|
| AVG | Hits ÷ at-bats | Walks, hit-by-pitches, extra-base power |
| OBP | (Hits + BB + HBP) ÷ (AB + BB + HBP + SF) | Power — a single and a double both just "get on base" |
| SLG | Total bases ÷ at-bats | Walks entirely — SLG ignores plate discipline |
| OPS | OBP + SLG | Weights OBP and SLG equally, though OBP correlates more strongly with runs scored |
A player who draws 100 walks a year is generating scoring opportunities that never touch the batting-average denominator. That's why a .250 hitter with a strong walk rate can outproduce a .300 hitter who never takes a pitch — batting average was never built to see the difference.
wOBA: weighting outcomes by what they're actually worth
On-base plus slugging (OPS) is an improvement, but it's still a blunt instrument — it just adds two rates together as if a point of OBP and a point of SLG contribute equally to runs, which they don't. wOBA (weighted on-base average) fixes that by assigning each outcome a coefficient derived from its actual run value, not a guess. FanGraphs' published 2013 weights, as an example of the method:
wOBA = (0.690×uBB + 0.722×HBP + 0.888×1B + 1.271×2B + 1.616×3B + 2.101×HR)
÷ (AB + BB − IBB + SF + HBP)Those coefficients aren't arbitrary — they come from measuring how many runs a league actually scored following each type of event, and they're recalculated every season as league-wide scoring levels shift. The output is scaled to read like OBP: a league-average wOBA sits around .320, so a .380 wOBA reads as clearly above average the same way a .380 OBP would, but it now reflects power and discipline in one number instead of needing OBP and SLG side by side.
OPS+ and wRC+: same idea, adjusted for context
A .900 OPS in 1968's pitcher-dominated season and a .900 OPS in a modern hitters' park are not the same accomplishment — league scoring levels and home ballparks both move the raw numbers around underneath a player who hasn't changed at all. OPS+ (and wRC+, built the same way on top of wOBA) solves this by indexing every player to their league and park context, with 100 set as league average: a 130 OPS+ means 30% better than a league-average hitter that year, in that park, full stop — which is why OPS+ is the number you'll see used to compare a player across seasons, or across eras, when the raw stats alone would be misleading.
Pitching has the same blind spot: ERA vs. FIP
ERA has the same flaw AVG does — it's heavily shaped by outcomes the pitcher doesn't fully control, like whether a routine grounder finds a gap or lands in a glove. FIP (fielding independent pitching) isolates the three outcomes a pitcher does control almost entirely — strikeouts, walks, and home runs — and ignores everything that depends on defense:
FIP = ((13 × HR) + (3 × (BB + HBP)) − (2 × K)) ÷ IP + FIP constantThe FIP constant is a league-wide adjustment that puts the result on the same numeric scale as ERA each season, so a FIP of 3.50 and an ERA of 3.50 mean the same thing at a glance — but a pitcher with a 4.50 ERA and a 3.20 FIP has been getting unlucky on balls in play, not pitching worse than the number suggests.
WAR: one number, two different formulas
WAR (Wins Above Replacement) tries to answer "how many more wins is this player worth than a freely available replacement-level player at the same position," folding hitting, pitching, fielding and baserunning into a single figure. The catch: Baseball-Reference's WAR and FanGraphs' WAR are not the same calculation and routinely disagree, most visibly on pitchers — FanGraphs' version (fWAR) builds pitcher value from FIP, while Baseball-Reference's version (bWAR) uses runs actually allowed per 9 innings (RA9). Quoting "5.2 WAR" without saying which source it came from is the single most common way this stat gets misused in an argument.
Common mistakes
- Comparing WAR values across sites without checking the source. A 6.0 fWAR season and a 6.0 bWAR season for a pitcher can reflect meaningfully different underlying performances.
- Reading OPS+ or wRC+ as a raw percentage of hits. It's already era- and park-adjusted; a 115 OPS+ two decades apart represents the same relative production, even if the raw OPS numbers behind them look nothing alike.
- Treating a low ERA as proof of a great pitcher in a small sample. A handful of starts with a favorable defense behind them can produce an ERA that FIP doesn't back up at all — check both before drawing a conclusion.
- Using batting average alone to settle a "who's better" argument. It answers one question — contact rate on official at-bats — and was never designed to capture power or plate discipline.
FAQ
Is batting average completely useless now?
No — it's still a fine measure of one specific thing, how often a hitter turns an at-bat into a hit. The mistake is treating it as a measure of overall offensive value, which it was never built to do. Pair it with OBP and SLG (or just look at OPS+/wOBA directly) for the fuller picture.
What counts as a "good" wOBA?
League-average wOBA tracks league-average OBP by design, so if you know roughly what a good OBP looks like in a given era, you know what a good wOBA looks like — commonly cited as around .320 for average, .370+ for good, and .400+ for an elite offensive season.
Why do WAR numbers for the same player differ between websites?
Because "WAR" isn't one formula — it's a family of models that agree on the concept (value over a replacement-level player) but differ in the components, most significantly how each site values pitching and defense. Always check whether a cited WAR figure is bWAR or fWAR before comparing it to another source.
Cover photo by Phil Evenden on Pexels.