SpaceCityAg05 said:

---

It is not just batting average that is an incomplete stat, but RBI and R are as well. All of those are partial indicators, which is what old school analysis looked at and drew conclusions off of. Analytic try to look at complete picture of offensive contribution, independent of teammates. RBI reflect the guys in front of you getting on base. R reflect guys behind you driving you in. Average indicates how often balls in play are turned into hits, but is blind to hit quality and how what you do when you don't put the ball in play.

Here is a comp:
Chandler Simpson
Slash lines: .305 AVG/.333 OBP/.353 SLG/.687 OPS
162 hits in 532 AB
188 Total Bases

Isaac Paredes
Slash lines: .260 AVG/.359 OBP/.435 SLG/.795 OPS
135 hits in 519 AB
226 Total Bases

I intentionally chose Paredes as a comp because he is a good hitter who is not an All-Star masher like a Yordan, Caminero, or a team's typical 2/3 hitter.

Chandler's average makes him look much better, but he only has 27 more hits than Isaac. With 12 games to play (I am going by team games played instead of individual here), Simpson gets 1.08 hits per game, while Isaac gets 0.9. Over the course of the season, Simpson gets one extra hit per week compared to Paredes.

On the flip side, Isaac has generated 38 more bases than Simpson. This is because he walks more (when not swinging the bat) and has better quality hits (SLG). This means he is on base more often to potentially be driven in and does more to drive in runners than Simpson. Now Simpson does add value with his SB, but each CS (14) offsets each SB (41) by MORE than 1 because an out on the base paths hurts more than a gained base adds.

The difference is actually bigger than it looks: the biggest flaw of OPS is that it double counts times on base. But that means it actually overrates singles/walks guys compared to sluggers. This is why wOBA is becoming more popular.

At the end of the day, Simpson is a good young player, but his WAR is defensive aided. He is a slightly below average offensive player despite hitting .300. Isaac is a high quality offensive player despite "only" hitting .260.

---

Well said, but I'd argue one minor point. OPS actually overweights the sluggers. Because SLG is, generally 25-50% higher than OBP, OPS is made up more of SLG than OBP.

To see this play out with a real-world example, compare Matt Olson and Yandy Diaz.
Olson: .323 OBP, .505 SLG, .828 OPS
Diaz: .366 OBP, .463 SLG, .829 OPS

Nearly Identical OPS, yet they get there in a very different way,

Olson: ..353 wOBA
Diaz: .360 wOBA

Diaz is less than .1% higher in OPS yet almost 2% better in wOBA.

OPS actually overrates batting average in isolation. This is because hitting singles will increase both factors (essentially double counted). Basically the "Hit" stat becomes overrated.

As a demonstration, imagine two equivalent (judged by OPS) batters.

Player 1's slashes are .300/.350/.400 and therefore has an OPS of .750
Player 2's slashes are .250/.350/.400 and therefore also has an OPS of .750

Almost every traditionally thinking baseball person would likely think that Player 1 is likely to be a better offensive player. However, counterintuitively, because of the double counting of hits that I mentioned, Player 2 is likely better. This is because Player 2 has to have walks and extra base hits to make up the difference, and those will be worth more than the extra singles hit by Player 1.

If I create the two players statistically, then over 600 plate appearance each, here are the lines used.

Player 1: 600 PA, 570 AB, 171 H, 30 BB, 133 singles, 23 doubles, 5 triples, 10 home runs
Player 2: 600 PA, 540 AB, 135 H, 60 BB, 91 singles, 23 doubles, 5 triples, 16 home runs

So by choosing Player 1 over 600 plate appearances, you would gain 42 singles, but lose 30 walks and 6 homeruns.

30 walks and 6 home runs is demonstrably better than 42 singles if you think about the outcomes. If you roughly weight a single and a walk equally, you are essentially taking 12 singles instead of 6 homeruns by choosing Player 1.

JJxvi said:

---

OPS actually overrates batting average in isolation. This is because hitting singles will increase both factors (essentially double counted). Basically the "Hit" stat becomes overrated.

As a demonstration, imagine two equivalent (judged by OPS) batters.

Player 1's slashes are .300/.350/.400 and therefore has an OPS of .750
Player 2's slashes are .250/.350/.400 and therefore also has an OPS of .750

Almost every traditionally thinking baseball person would likely think that Player 1 is likely to be a better offensive player. However, counterintuitively, because of the double counting of hits that I mentioned, Player 2 is likely better. This is because Player 2 has to have walks and extra base hits to make up the difference, and those will be worth more than the extra singles hit by Player 1.

If I create the two players statistically, then over 600 plate appearance each, here are the lines used.

Player 1: 600 PA, 570 AB, 171 H, 30 BB, 133 singles, 23 doubles, 5 triples, 10 home runs
Player 2: 600 PA, 540 AB, 135 H, 60 BB, 91 singles, 23 doubles, 5 triples, 16 home runs

So by choosing Player 1 over 600 plate appearances, you would gain 42 singles, but lose 30 walks and 6 homeruns.

30 walks and 6 home runs is demonstrably better than 42 singles if you think about the outcomes. If you roughly weight a single and a walk equally, you are essentially taking 12 singles instead of 6 homeruns by choosing Player 1.

---

Not quite true.

https://www.fangraphs.com/tools/guts

Based on 2026 Data
HR = 2.049
1B = 0.890
BB = 0.698

6 HR + 30 BB = 33.234
42 1B = 37.38

I got these from AI so I apologize if they're not 100% accurate, but I wanted to have some type of # for this exercise.

25% of the time a runner on first scores
14% of the time a single drives in a run
2% of the time a walk drives in a run
The average HR drives in 1.57 runs

The 42 singles guy drives in 10.5 runs and scores 5.88 runs = 16.38 Runs
The 30 walks and 6 HR guy drives in (0.6+9.42) 10.02 runs and scores 7.5 * = 17.52 Runs

*only counting runner on first scoring since the RBI and run scored on a HR isn't double counted.

This does not account for the benefit of moving runners into better positions. A Single can move someone from 1st to 3rd or 2nd to 3rd where a walk cannot.

That just may make up for the difference in the Runs. Or maybe AI isn't as spot on.

Regardless, the 2 situations are very similar, and one isn't demonstrably better.

This World Series was interesting for the analytics vs old-school/anti-analytics crowd.

Antis claiming the Brewers & Rays were going against analytics while both teams have huge analytics staffs and are well known for diving deep into them.

I've been critical myself of the worshipping of analytics but I think what this Series showed is that it's not necessarily analytics themselves but looking at a lot of the numbers and drawing the wrong conclusions and formulating the wrong offensive strategy and identity for your personnel with what the numbers tell you.

Basically, both teams dive deeply into the numbers and interpreting them differently than most of the rest of MLB. Or perhaps their staff is looking at the numbers and saying the conventional wisdom on how to use them is incorrect.

We all know there is more than 1 way to score runs. These teams operate on a tiny budget when compared to the Mets, Dodgers, etc. They have to find runs cheaply. There was a time you could do that with HRs from guys who did other stuff ugly. Now it seems the pendulum has shifted. We have automated strikes now. Let's load up on short guys to draw walks. Fielders have gotten stupid good at positioning. Let's start working the opposite field. 10 years ago, X exit velo and Y launch angle meant doubles Z% of the time. Outfielders are faster and positioned to take those away. Paying for guys with the proper X/Y EV & LA isn't as worth it. And so on.

I think you're looking at analytics wrong if you are concluding that the numbers ever told you that there was only one way to do things. The analytics departments for these teams just try to sift through all of the different data points for guys on their own teams and prospects/free agents to figure out how to best translate what those guys do into value.

The three true outcomes side of things sucked up a lot of the air in analytics conversations among fans, but the data has also been telling us consistently that the Brice Turangs, Nico Hoerners and Steven Kwans of the world are 5 WAR players.