New Year’s Holiday Sale: ends december 31st
Before we throw weighted balls under the bus, let's talk about what this study actually measured. Comparing raw torque values across different athletes without normalizing for body weight or individual characteristics is like comparing how much two people bench press without knowing one weighs 150 pounds and the other weighs 220. The numbers don't tell you what you think they do.
A 2020 study presented at the American College of Sports Medicine examined 21 collegiate pitchers (13 who trained with weighted baseballs and 9 who didn't) to determine whether weighted ball training increases performance or injury risk. Researchers measured throwing velocity and maximum elbow valgus torque during 15 fastballs per pitcher, comparing the two groups to see if weighted ball training created a performance advantage or elevated joint stress. The results? Weighted ball trained pitchers threw with more than double the elbow torque (110 N⋅m versus 52 N⋅m), but showed no significant difference in velocity (36 m/s versus 35 m/s). On the surface, this looks damning. But here's the problem: this study has some serious limitations that we need to address before we start making sweeping claims about weighted balls.
The weighted ball group averaged 36 meters per second (roughly 80.5 mph), while the non-weighted ball group threw 35 meters per second (about 78.3 mph). Statistically, that's not a meaningful difference. So if the velocity gains weren't there, what's the issue? The torque. Weighted ball trained pitchers experienced 110 Newton-meters of elbow valgus torque compared to just 52 Newton-meters in the non-weighted ball group. That's more than double the stress. At first glance, you'd think this proves weighted balls wreck elbows. But let's pump the brakes.
First, this was just an abstract that never became a full article. But even setting that aside, the real problem is how torque was measured. The increased elbow torque we're seeing is likely a result of adaptations to the training, specifically increased range of motion at layback or scapular retraction. Deeper ranges mean more time to produce force, which can result in higher torque measurements. But here's the critical issue: torque, unless normalized for body weight, can basically be irrelevant.
To be honest, this reminds me of a 2025 study that analyzed 624 collegiate and professional pitchers. The researchers found that body weight showed the strongest correlation with elbow torque, with each 2.25-pound increase resulting in a 1 Newton-meter increase in torque. That's massive because it tells us heavier athletes will naturally generate higher absolute torque values. So when we're comparing a 220-pound pitcher who trained with weighted balls to a 175-pound pitcher who didn't, we're not just comparing training methods. We're comparing two completely different physical systems.
A 2019 study examining variability in throwing metrics explicitly stated that velocity is a more effective predictor of effort than torque when torque is not normalized to body weight. Without normalization, a 200-pound pitcher and a 160-pound pitcher throwing the same 75 mph could generate very different torque values, yet most studies treat them equally.
But it gets even more interesting. A 2021 study looked at 91 professional pitchers and compared how velocity relates to joint stress within individual pitchers versus across different pitchers. The finding? Across different pitchers, ball velocity only weakly predicted elbow stress (R² = 0.175). But within a single pitcher, velocity strongly correlated with elbow varus torque (R² > 0.85). Translation: comparing raw torque values between two different pitchers tells you almost nothing about their relative injury risk.
And here's where it gets really good. A 2022 study investigated weighted baseballs (85 to 198 grams) in 26 collegiate and professional pitchers experienced with weighted ball training. The result? No significant differences in elbow varus torque were found across ball weights. Experienced throwers showed no increase in joint stress despite using implements ranging from underweight to significantly overweight.
So what's going on? A 2025 study examined how physical characteristics influence the torque-velocity relationship. The researchers found that shoulder strength, mobility, grip strength, bodyweight, and trunk control all altered how much elbow varus torque pitchers experienced for a given velocity. Greater shoulder internal rotation strength lowered torque. Greater shoulder flexion range lowered torque. But higher bodyweight and stronger lead leg stability increased torque (likely because they enable more force generation). The lesson? Elbow stress isn't simply a byproduct of velocity. The whole kinetic chain shapes it. And that means torque is not universally comparable.
This matters because we're making decisions about training tools based on fundamentally flawed comparisons. When you see a study claiming weighted ball training doubles elbow torque, you need to ask: did they normalize for body weight? Did they account for range of motion adaptations? Did they consider the physical characteristics that influence torque-velocity relationships? In most cases, the answer is no.
Torque is not universally comparable. A 180-pound high school pitcher generating 60 Newton-meters at 80 mph is experiencing very different relative stress than a 220-pound professional generating 100 Newton-meters at 95 mph. But if you just look at raw numbers, you'd think the pro is at much higher risk. You'd be wrong. The pro might actually be handling that load better relative to their strength, tissue capacity, and mechanical efficiency.
What's also severely lacking is any detail about the protocols. When they say "trained with weighted balls," what does that actually mean? How often? What weights? What progressions? The weighted ball is just a tool. Application is what matters. Perhaps several of these athletes didn't need weighted balls as an implement in their program. Just like torque isn't universal, neither should any tool or program be applied that way.
First, stop comparing your torque to someone else's torque unless those values are normalized for bodyweight and velocity. If you're monitoring elbow stress, track changes in your own torque relative to your own velocity. Are you throwing harder with the same torque? That's efficiency. Are you generating more torque for the same velocity? That's a red flag that something in your kinetic chain might be breaking down.
Second, if you're using weighted balls (or any training tool), make sure the application matches the need. Are you addressing a specific constraint? Building arm speed? Improving sequencing? Or are you just doing it because someone told you it works? The tool is neutral. The outcome depends on how and why you're using it.
Third, understand that increased range of motion and higher torque aren't inherently bad. Professional pitchers often generate significantly more torque than high school pitchers, not because they're at higher risk, but because they've built the strength and sequencing to handle that stress. If you're going to train in deeper ranges (which weighted balls can facilitate), you need to make sure your tissues are prepared for the demand. That means targeted strength work and respecting recovery periods.
Fourth, recognize that individual variation is massive. Some athletes will respond incredibly well to weighted ball training. Others won't. The best programs are the ones that adapt to the athlete, not the ones that force every athlete into the same mold.
Are we vilifying a tool because of poor application, or are we genuinely identifying a risk that exists regardless of how it's used? Based on the research, it's pretty clear that the problem isn't weighted balls. The problem is how we measure and interpret stress. Raw torque comparisons without normalization are essentially meaningless. They don't account for bodyweight, they don't account for mechanical efficiency, and they don't account for individual physical characteristics that dramatically influence the torque-velocity relationship.
The Aguinaldo study showed that weighted ball trained pitchers threw with higher torque but no velocity advantage. But that doesn't prove weighted balls are dangerous. It proves that in this specific population, with this specific application (whatever that was, since we don't actually know the details), the training didn't produce the intended velocity outcome. Maybe the program was poorly designed. Maybe the athletes weren't ready for that training stress. Maybe they needed different programming altogether. We don't know. What we do know is that experienced throwers using weighted balls under appropriate progressions show no increase in joint kinetics. And we know that torque varies massively between athletes based on bodyweight, strength, mobility, and sequencing.
So the next time you see a study claiming a tool is inherently dangerous, dig deeper. Ask whether the comparisons are valid. Ask whether the application was appropriate. And most importantly, remember that your body is unique. What works for someone else might not work for you. What stresses someone else's elbow might not stress yours the same way. Train smart, track your individual responses, and stop trying to fit every athlete into the same box. a risk that exists regardless of how it's used? Based on the research, it's pretty clear that the problem isn't weighted balls. The problem is how we measure and interpret stress. Raw torque comparisons without normalization are essentially meaningless. They don't account for bodyweight, they don't account for mechanical efficiency, and they don't account for individual physical characteristics that dramatically influence the torque-velocity relationship.
The Aguinaldo study showed that weighted ball trained pitchers threw with higher torque but no velocity advantage. But that doesn't prove weighted balls are dangerous. It proves that in this specific population, with this specific application (whatever that was, since we don't actually know the details), the training didn't produce the intended velocity outcome. Maybe the program was poorly designed. Maybe the athletes weren't ready for that training stress. Maybe they needed different programming altogether. We don't know. What we do know is that experienced throwers using weighted balls under appropriate progressions show no increase in joint kinetics. And we know that torque varies massively between athletes based on bodyweight, strength, mobility, and sequencing.
So the next time you see a study claiming a tool is inherently dangerous, dig deeper. Ask whether the comparisons are valid. Ask whether the application was appropriate. And most importantly, remember that your body is unique. What works for someone else might not work for you. What stresses someone else's elbow might not stress yours the same way. Train smart, track your individual responses, and stop trying to fit every athlete into the same box.
.png)
.png)
.png)
