@kagami

expanding on passan’s post to breakdown the math, moneyball, and rise of advanced metrics 1. the math avg = hits / at bats obp = (hits + walks + hit by pitches) / (hits + walks + hit by pitches + sacrifices) slg = total bases / at bats so here’s the thing about batting average, it makes some key assumptions: - all hits are created equal (a single = a homer) - walks and hit by pitches dont matter (at bats = plate appearances - walks - hit by pitches) - “luck” is a non-factor when hitting but we know that a homer is more valuable than a triple which is more valuable than a double which is more valuable than a single, right? and walks matter are valuable. in fact, walks are a more durable skill than getting hits (which have a luck based component). 2. the moneyball era while most people think about brad pitt and finding undervalued talent to field a competitive team of a small budget, moneyball is really about valuing players by obp and slg over avg. the a’s philosophy was that players with high obp and slg were more valuable than high avg players. essentially, the a’s preferred players who walked a higher rates to offset a lower avg. and, despite a low avg, when these players got hits, it was more likely to be an extra base hit (which we know is more valuable than a single). this is probably easier to illustrate, so let’s look at two players: player a: .298 / .333 / .407 player b: .236 / .365 / .463 if we just look at triple slashes (avg / obp / slg), we can see that player a hits for high average, but to hits are more likely to be singles or doubles. player b, despite a much lower average, gets on base more and hits for more power. in a vacuum, why wouldn’t you want player b? since a walk is the same value as a single and player a hits mostly singles and doubles, player b is more valuable with an obp > player a’s avg and player b hits for more power. 3. the rise of weight on base average (woba) while obp + slg improved on measuring players value, the metrics made a few key assumptions: - obp still assumes all hits are created equal - slg assumes that a double is 2x as valuable a single, a homer is 4x as valuable as a single, etc woba takes this even further by calculating the expected run value of each way to get on base adjusted every year. for 2025, a homer is worth 2.053 in run value while a walk is worth 0.692 in run value. in essence, woba reflects the run value a player creates per plate appearance. so, let’s revisit player a and player b and add the context of woba. player a: .298 / .333 / .407 / .324 player b: .236 / .365 / .463 / .351 so, we can see why avg has fallen out of favor. a high average player that lacks power and on-base skills creates less run value per plate appearance than a player with low average that possesses power and on-base skills. this shift from avg -> obp/slg -> woba is also reflective of why we’re in a baseball era where teams prioritize power over contact leading to less balls in play and high strikeout rates. https://x.com/jeffpassan/status/1951747434549289092?s=46&t=dGdml54Bn5gAjlxSoZTivg

2 replies

1 recast

9 reactions