Let’s discuss something that Ben Clemens of FanGraphs posted recently. He took a tweet from Tom Tango, which asked his readers which one they would prefer:
A. .315/.365/.510
B. .265/.365/.510
Implied is that these two slash lines are recorded in the same year/park, or in different eras/parks but with the same run-scoring environment. Because putting up one line in Coors Field in 2000 is a lot different than putting up the other line in Dodger Stadium in 1965.
Most people instinctively pick A. There’s a chance to do more damage with a hit than a walk. But it’s a little bit like asking which one weighs more – 10 pounds of concrete or 10 pounds of feathers.
Clemens goes thru a lot of math and if that stuff interests you, make sure to click the link at the top of this piece which will bring you to the original article. If you’d rather skip the details, the takeaway is this:
The modern view that OBP and SLG are more important than batting average for scoring runs isn’t opinion or preference. It’s borne out by the way that real teams score real runs in real games. Batting average is better than nothing, but it’s meaningfully worse than the other statistics we have available, and adds no useful information if you already know OBP and SLG.
From the example above, we know immediately that Player A gets more hits. We figure out pretty quickly that Player B gets more BB/HBP. But what may not jump immediately to mind is that Player B gets more bang for his buck with his hits. Player A has a .195 ISO – which is really good. But Player B has a .245 ISO, which is great.
In the comments section, someone says that in 600 PA, it works out to about 47 singles for Player A, compared to 40 BB and 7 HR for Player B. Another commenter chimed in that the difference in wOBA would be .001 in favor of Player B.
We use OPS and OPS+ a lot here, despite it being a sort of crime to add two things with different denominators. We do it because it’s easy and it gets us 95% or so to the right answer. There are things upon which reasonable people can disagree. But using AVG instead of OBP or OPS or OPS+ or wOBA or wRC+ is simply not one of them.
Like many of you, I grew up when AVG was king. If you ask me what number sticks out most for a player in my Mets fandom, the response – easily – is Cleon Jones’ .340 AVG in 1969. But as adults, we have to recognize that things drilled into our heads as kids aren’t necessarily true. And this is the case with real life, as well as sports. No, George Washington Carver didn’t invent peanut butter. And also no is the idea that batting average is the best way to judge players. Or that it gives you better information than other stats.
I resonated with your comment about growing up with different stats and/or different estimations of the value of certain stats. I still want to make RBIs significant, although I know that’s relative to the quality of your teammates. I never was any good at math in school and I easily get lost in the new stats. I have no idea what ISO is and, frankly, please don’t try to explain. I wouldn’t understand it and even if I did today, I wouldn’t remember its meaning tomorrow, lol! I’ll just rely on all of you on Mets 360 to discuss what’s good and what isn’t.
ISO is Isolated Power. Subtract AVG from SLG and that’s ISO. Babe Ruth is the all-time leader at an other-worldly .348, Mike Trout is the active leader at .284. Of those who played for the Mets, the career ISO leader is Carlos Delgado (.266). Shows you bang for your buck with hits in a way that’s better than SLG.
I still think in terms of batting average, home runs and rbi’s. That is what happens when you spend countless hours as a kid reading the back of baseball cards and a 1975 baseball encyclopedia.
This morning I was looking at career OPS+. I was astonished to see that Frank Howard had a 142 career OPS+, which is higher than many hall of famers.
I ran into that article and tried reading it, but gave up halfway through. Just too wordy to describe exactly what you wrote in a few sentences. I’m a big believer that a hit is not as good as a walk and like WPA because it agrees. It gives a single a value of 1, but a walk a value of 0.78. I think that’s fair.
However, when ISO+ 50 points higher, you need to take notice. When Keith Hernandez used to make fun of Duda’s “clutch walks”, there was a reason. Need to go to the plate looking to do damage, otherwise why bring a bat?
The article is quite aggravating and certainly exposes how searching for the answer you want even with math will allow you to find what you are looking for. The initial premise of the question and the subsequent data are unrelated to each other. The posit brought by Tango asks about the productivity of a single player triple slash. Going forward, the data show team stats. So whatever comparison you get is totally mistaken. I’d take player 100% of the time, because the person with the higher BA is going to drive in more runs than the person that walks every time, or who might advance on an error. So while both players have the same OPS, and so are valued equally, only one player is hitting the ball, and therefore creating more *team* opportunity.
It is these exact kinds of slight-of-hand discussions that really cloud the use of stats and makes me queasy about the flexibility of judging individuals based on a team analysis.
You’re completely ignoring the SLG advantage of the other player. Maybe those singles drive in a run. We know those HR do.
Nowhere does it say that the batters are coming up in a specific situation. So, we have to think these are all “average” situations. Linear Weights show the average value of a BB as .29, a single as .44 and a HR as 1.39
So, if it’s 47 singles versus 40 walks and 7 HR it’s
47 * .44 = 20.68
(40 * .29) + (7 * 1.39) = 11.6 + 9.73 = 21.33
Question re: the linear weights. Are the units relevant? Player B has a 0.65 advantage in linear weights over Player A. 0.65 what? Runs? Over the course of a full season, this doesn’t amount to much. On the other hand, maybe these units don’t correspond to anything. The advantage is a mere, pi-like 3.14%. That’s not much.
For me, being old school, I appreciate BA and like to see the ball in play, where many good things can happen. A 12 pitch walk is fantastic too. But IMHO, the cost of a K is understated and I will never be a fan of the three true outcome hitter.
I will take Player A all day long and twice on Sunday.
Please keep in mind the title of the article.
Yes, Linear Weights is measured in runs. And that was brought up in response to ChrisF, who said that he’d take one player 100% of the time. Now, from a personal preference, it’s certainly ok to prefer one over the other. But if you’re concentrating on doing what it takes to push the team forward – there’s no significant advantage in either player.
I can appreciate player B with a higher ISO, but I would need a further explanation as to why that player would we preferred to another with an identical OBP SLG OPS. My quick conclusion would be that the players provide equal value to a team. Even if one hitter may be feared by the opposing pitcher more than others.
That’s just it – both players are fairly even. Maybe you can find a fraction of a run to prefer one over the other. But there’s no clear cut preference.