Ideas about why this offense was so inept at scoring. And that pesky boxplot returns. What is with that?
One day late in the 2024 season, the Cardinals were experiencing another miserable day getting runs home. They ended up a whopping 2-for-19 batting with Runners In Scoring Position (RISP) and lost 4-3 in extra innings to Toronto.
Of course, we’d seen this all season. Not to the extreme of 2-19, but the RISP performance was consistently bad, seemingly throughout the season and across the whole line-up. On the boards, some opined it was an approach issue and recommended firing the hitting coach (done!). Others patiently reminded us of all the studies that show that RISP is random. That it regresses to the player’s overall mean over time, and it is not predictive in any way. A player who has gone 5-7 with RISP (looking at you, Ohtani) is not a shoe-in to get a hit in the next AB with RISP. But if it is so random, and not a good predictor of future performance, how is it we all could predict the outcome of innings with eerie accuracy? Goldy would K. Arenado would pop out. Burleson would roll over and ground the ball to the first baseman. Over and over this happened. And I got curious.
No, I did not set out to prove the studies wrong. RISP is random and is not predictive. However, I wanted to know what WAS going on. I started off with a basic question … if it is NOT the hitting coach’s fault (he took a grand total of ZERO of the RISP at-bats this year), what is it? Here is what I found.
First, I started off wanting to know how the Cardinal’s line-up as a whole stacked up against the rest of the league. So, I wrote some queries and examined every batted ball event in 2024, looking at various advanced metrics.
In ABs with no runners on, the Cardinal offense was middle of the road across those metrics. Upper half in OBP, K-rate and wOBA (which directly relates to runs). But they were lower half in ISO and walk rate. This made them middling, confirmed when their wRC+ placed them smack-dab in the middle at 15. What does low ISO and much higher OBP tell us? They hit a lot of singles with the bases empty.
Then I looked at situations with runners on (scoring position or not). To be fair, the Cardinals offense improved when runners were on, in terms of raw metrics. I bet that surprises many! OBP up 13 points. Walk rate up a full 1%. wOBA up 6 points. But look at all those rankings. They all lowered (except walk rate). The rankings, not the raw scores. What does this tell us? The batters were pretty much the same with runners on as they were with no one on, maybe a little better. Except, the rest of the league hits a lot better with runners on than with no one on. What gives?
First, look at that ISO. .144 is … not great. In fact, it’s pretty poor (20th overall). Not enough doubles, triples and homers. They ranked 23rd in homers, 17th in doubles and 17th in triples. No power, no speed. This is a team that strung together singles like crazy (5th in the majors) and had trouble scoring.
With that context, I then looked exclusively at RISP events. The Cardinal results are not that far off their norms with no runners on, except their ISO plummeted. OBP was 3 points different. K-rate .7% worse. Walk rate still better (1% above runners on, 2% above no runners on). wOBA plummeted along with ISO. I think that is not a coincidence, as ISO weighting factors favor doubles, triples and homers, which the Cardinals did not specialize in. As a result, wRC+ plummeted from middle of the road with no runners on to bottom of the league with RISP.
Let’s look at some individual results (wOBA) compared to league, in any situation. Below, you will see how the Cardinals performed (red dots), overlaying a 4-quartile box plot for the rest of the league. 4 Cardinal hitters in the top quartile, 4 guys in the 2nd quartile, 5 guys in the third and 6 in the fourth. So, a tad below average. No surprise there.
NOTE: The mere mortals comment means that I had to take out a unicorn result (Ohtani) to make the boxplot more readable. His outsized wOBA really stretched the diagram otherwise.
Those six in the lowest quartile? Fermin (lowest), Crawford, Saggese, Scott, Siani and Walker. Anyone curious about the 4 in the highest quartile? Contreras, Herrera, Donovan, Nootbar.
Looking above, measured by wOBA, everyone spreads across the bell curve pretty neatly, skewed a little bit toward the bad side, but a normal distribution. But the Cardinals weren’t really an average to slightly below average offense. They were bad, if you consider 22nd out of 30 to be bad.
What gives? Well, look at the next slide, where I narrow to RISP.
If you look carefully, you’ll notice that the Cardinal results pretty much stayed static. What moved is the underlying boxplot. The quartiles. Because the league average wOBA increases a fair amount between no runners on and RISP. Except in St. Louis. In St. Louis, they remain the same hitters, whereas everyone else improves. Why?
Let’s look further. Change the plot from wOBA to ISO. What happens? The blue dots are the outliers at the top of the spectrum. Judge, Ohtani, Seager, Harper, and …. Tyler O’Neil. Too funny! The Cardinals pretty much all fall in line way down in the 4th quartile. To me, wOBA draws a straight line to runs. ISO draws a straight line to power. What this chart says to me is “No power, no runs”.
But inquiring minds might want to know, did the Cardinals hit into back luck with BABIP? Let’s check. Nope. League average BABIP .273. Cardinal average BABIP .274. Bet not many people would have guessed that one. The distribution shown below as:
Referring to the very first chart in this article (repeated below), the Cardinals ISO drops from .144 overall (bad) to .113 w/ RISP (horrible), while the rest of the league rises from .113 to .190. Think about that. Almost half the power events of the rest of the league (ie. the average), not just the top hitting teams.
From all this, I conclude that a below average-ish hitting team with little power hit below-average-ish with RISP with even less power.
So, I’d make the case that this isn’t an approach problem, it is a talent problem. A team with consistently low walk rates and ISO production that ranges from bad to horrible in all situations is not going to score very much.
I’d go a step further and say this team needs to add OBP and stack it at the top (instead of the pair of .314 OBP batters they had there the majority of 2024) and they need to get some power (ISO) into the 3-4-5 holes, and add something down the line-up against LHP to offset Siani’s sub .600 OPS, whether that comes in slugging or OBP, I don’t think it matters.
Turns out, the sages were right. RISP results do regress to the mean talent level. 2024 was not a random outcome. The Cardinals performed to their talent level. The combination of average-ish BsR, poor walk rates and bad to horrible ISO doomed this offense.