Foolish Bailey Took the Pop Up Pill, Should You?

Over the weekend, I watched Foolish Bailey’s most recent video on pitcher command. He used Red Sox rookie starter Jake Bennett to run through the different ways to quantify command.

It’s pretty fun. 

In it, Bailey points out that Bennett intentionally throws his fastball up in the zone, using Bennett’s heat map to demonstrate frequency, his 115 Location+ to demonstrate efficacy, and Bailey’s own “Upstairs” metric, describing pitches thrown just above the strike zone, to reiterate the heatmap and emphasize Bennett’s intent relative to other pitchers, as Bennett leads all pitchers on percentage of Upstairs four-seamers this season.

Bailey went on to highlight Bennett’s lackluster four seam fastball in a vacuum, that is elevated by tunnelling with and playing off of his other pitches, particularly his sinker and changeup. All three of these pitches, Bailey points out, have positive run values for the rookie. 

But what was most interesting to me was how Bailey concluded the video. He’d already proven Bennett throws with intent, locating his fastball well in relation to his offerings with better stuff to maximize his outcomes. But then he shifted to batted ball outcomes. Bennet’s sinker has elicited a 69% groundball rate and his fourseam a 26.7% IFFB%.

Bailey described himself as having “taken the pop up pill”, comparing infield pop ups to strikeouts in terms of utility to a pitcher. He showed us a list of pitchers with 70+ innings this season and both high GB% and IFFB%. Here’s that list. 

Obviously, there’s some aces here and some low ERAs. 

Bailey then shows Bennett’s league-leading Launch Angle Sweet Spot rate, which places him atop a list that includes Shohei Ohtani, Trevor McDonald, Janson Junk, and Framber Valdez. Not as impressive as the other list. 

The premise seemed logical to me. As Bailey puts it, pitchers with good command throw pitches in locations that get a lot of strikes and a lot of suboptimal contact. 

I was left wanting more. I wanted to know if Jake Bennett possesses a skill that historically is repeatable for pitchers. I think we assume we know “ground ball guys.” But is inducing infield pop ups a repeatable skill? And does it impact ERA in the same way a high ground ball rate does? I wanted to know if Bailey had shown why Bennett’s ERA was great this season, and if that “why” was worth turning into a strategy to predict future season ERA for other pitchers. 

Here’s what I found. 

Does IFFB% Help Explain a Low ERA?

There’s a weak relationship between a high IFFB% and low ERA this season, but it’s not exactly evidence of Bailey’s optimal contact suppression creating aces.

But what is one season? Just a relative spoonful of a sample size. So I took all of the 100+ IP seasons from 2002 to today (since 2002 is the first season with IFFB% on Fangraphs). This gave me 3,292 pitcher seasons to work with. There was still a negative relationship between ERA and IFFB%, but it was even weaker. And clearly, there was year-to-year variation in the correlation between the two. 

So within the same season, IFFB% has little effect on ERA historically, though it certainly has so far this season. At this point, I’m starting to worry about Jake Bennett. And to answer your question, IFFB% has an even slimmer effect on next season ERA. 

What about IFFB% and GB%? Do these poor contact metrics (and others like it) correlate with low ERAs? 

Where Bailey stopped with IFFB% and GB%, I brought in more of the stats you can find on Fangraphs that describe contact, some familiar faces like HR/FB%, Hard%, Soft% to join GB%, as well as some command metrics like K-BB%, Zone%, Z-Contact%, and O-Swing%. 

As displayed above, HR/FB, K-BB%, and GB% had the biggest impact on in-season ERA, and an R² of 0.605 overall for the model demonstrated how much clearer a picture you can find using more than just IFFB% and GB% as proxies for weak contact and command. The model also did a better job than ERA at predicting next season ERA. 

It’s also worth noting that more than 1,000 of the 3,292 pitcher-seasons were followed by a pitcher-season of less than 100 innings, a reminder that the injury bug is always buzzing. 

So IFFB% matters a little. And GB% more so. Do these, and other stats in the model, stay sticky year over year? Are they measures of repeatable skills?

As I mentioned earlier, it feels like ground ball guys exist as a genre of pitcher, and particularly starting pitcher. Command artists are a genre too, one that is perhaps given less weight than ground ball rate in fantasy circles, because it’s connected to flukiness. 

For most pitchers, that is certainly the case. Earlier this season, I looked into starting pitchers running hot Location+ seasons. Between 2025 and 2026, none of these pitchers repeated, but some did have strong seasons during their Location+ run. It just did not look especially predictive of future Location+ or future ERA. 

But are some of the key metrics in this model for predicting ERA — K-BB%, HR/FB, and GB% — sticking as repeatable skills, season after season? 

It seems that metrics tied to quality of contact are far less sticky than command- and stuff-based metrics, with the exception of GB%. 

The blue graph on the left shows that amongst the more than 3,000 pitcher seasons of 100+ innings since 2002, GB%, K-BB%, Zone%, and Z-Contact% are the most consistent for a pitcher. On the other hand, there’s lots of year-to-year variance amongst pitchers on quality of contact and specifically, on the result of a fly ball, whether it’s an infield pop up or a home run. 

For fantasy analysis, this tells us that over a full season perhaps, and definitely across seasons, the ability to throw strikes and avoid contact on pitches in the zone are repeatable skills, and ones that do impact ERA, as shown earlier. Getting ground balls and getting more strikeouts than walks are also reliable from season to season, and again are strong ERA predictors. 

When specifically analyzing rookie starting pitchers, like Bennett, and when trying to see if a low ERA in a rookie season is repeatable, it seems to be worthwhile to focus on these measures, as opposed to just taking, say, rookie ERA at face value as predictive of next season.

The red graph on the right shows that there is a strong next season correlation for the core four statistics I highlighted: Zone%, Z-Contact%, K-BB%, and GB%, further emphasizing the ICC finding. 

Back to Bailey’s muse, Jake Bennett.

How does he look so far across these metrics. We already know he’s an ERA standout, and an IFFB% standout on his four-seamer at least. And his sinker, thankfully, induces ground balls. But does he throw strikes? And avoid zone contact? And does his command help him get more strikeouts and fewer walks?

This season, Bennett’s been slightly better than the average starting pitcher. And in those most repeatable metrics, the sticky icky, to quote Snoop, suggest that at the very least Bennett will be a strong ground ball guy, with at least league average strikeouts, Zone%, and Z-Contact%. 

And, as Bailey noted, Bennett is being intentional with his pitch mix and seems to be able to locate to maximize his arsenal, so there’s a chance that he can keep running his low ERA and even lower WHIP to help fantasy managers in his sophomore season. 

To wrap things up, I wanted to see if those four stickiest metrics, Zone%, Z-Contact%, K-BB%, and GB%, were correlated with success for pitchers this season. To compare across these metrics, I took their z-scores to standardize the comparison, and then weighted each based on my aforementioned model of relative correlation to ERA. 

Here’s the top 20 starting pitchers, with a minimum of 100 innings pitched this season, on these four key measures.

I’ll look back at this in the offseason, but I think I want to continue this line of analysis, emphasizing the stickiest, most repeatable metrics as skills that can be drafted around with more confidence than previous season ERA or WHIP, especially for young pitchers without a deep track record to outweigh peripherals. 

For now, I’ll just pick up Jake Bennett for his start against the Giants this weekend.