Mini Study: No Clear Winner as Bench Press, Push Press, and Military Press Compete for “King” of Upper-Body Lifts – MTI to Redo the Study

 

By Jackson Mann

BLUF (Bottom Line Up Front)

Four MTI Lab Rats completed a 5-week mini-study designed to explore which upper-body pressing movement provides the greatest strength transfer across horizontal and vertical patterns. Each athlete trained one lift—Bench Press, Push Press, or Military Press—using a 3x/week density progression, then tested for strength changes across all three lifts.

Results were mixed. While the Bench Press showed strong potential for overall transfer in some athletes, others experienced regression despite training the same lift. The Push Press yielded short-term gains that did not always hold. Military Press training failed to sustain strength across pressing patterns. Based on this data, no single lift clearly emerged as the most transferable pressing movement. 

We will redo the mini-study for a future geek cycle to see if there is a clear “king” of upper body lifts as the results were muddled at best. We’ll aim to tighten study controls, expand the sample size, and isolate pressing progressions from competing training demands to better understand true strength transfer between lifts.

Key Outcomes:

• Emmett (trained Bench): +15 lbs Bench, +15 lbs Military
• Jackson (trained Push): +15 lbs Push, 0 lbs Military, 0 lbs Bench
• Sam (trained Military): –15 lbs Bench, –20 lbs Push, –5 lbs Military
• Seung (trained Bench): +5 lbs Bench, –40 lbs Push, –10 lbs Military

Background

Upper-body pressing strength is essential for tactical athletes, mountain performance, and general physical preparedness. Most strength programs prioritize one primary lift—Bench Press, Push Press, or Military Press—but clear evidence on which offers the best carryover is lacking. This study aimed to evaluate which pressing movement delivers the most comprehensive transfer across the upper-body pressing spectrum.

Study Design

Participants

Four MTI Lab Rats were assigned to train a single upper-body pressing movement over a 5-week mini-study, with the goal of assessing which lift—Bench Press, Push Press, or Military Press—provided the most strength transfer across all three major pressing patterns:

• Emmett: Bench Press
• Seung: Bench Press
• Jackson: Push Press
• Sam: Military Press

Initial & Final Assessments

Each athlete completed 1-rep max (1RM) testing in the following lifts at both the start and end of the cycle:

• Bench Press
• Push Press
• Military Press

This allowed for cross-comparison of strength gains or losses in both the trained lift and the untrained lifts.

Strength Progression 

All athletes followed MTI’s density strength progression three times per week, with each session targeting only the athlete’s assigned lift. MTI’s density format is designed to build working strength, emphasizing time-restricted volume.

Each session followed this structure:

• 5 total rounds, performed on a set interval:
  • Every 90 seconds initially
  • Later adjusted to every 2 minutes after mid-cycle reassessment to accommodate rising intensity and allow for adequate recovery.
• 4 reps per round, using a progression of increasing percentages over the course of the cycle, restarting after the midcycle:
  • Sessions 1-2 at 80% of 1RM
  • Sessions 3-4 at 85% of 1RM
  • Session 5 at 90% of 1RM

Example Session:

• 5 Rounds of 4 Reps at 85% of 1RM Bench Press
  Every 90 Seconds

Results

Discussion

The results suggest that no single pressing lift consistently transfers strength across all pressing planes. Emmett, who trained the Bench Press, improved both his Bench and Military Press by 15 lbs, pointing to strong potential crossover. However, Seung—on the same Bench Press progression—gained only 5 lbs on the Bench and lost 40 lbs on the Push Press and 10 lbs on the Military Press. Seung mentioned that this drop in performance was most likely due to fatigue accumulated during the density cycle used to progress the bench press.

Jackson, training the Push Press, initially gained across all pressing movements at the mid-cycle mark but failed to maintain progress in the Bench and Military Press by the end of the cycle. Only his Push Press gain held (+15 lbs). This pattern may reflect his young training age—responding well early, but unable to sustain gains under cumulative fatigue.

Sam’s Military Press progression failed to deliver across any movement. Despite small mid-cycle gains, he ultimately regressed in all three lifts. Anecdotally, Sam noted that his shoulders were not feeling fully recovered from the previous training session which may have affected his final results. The overall drop in his performance can be attributed to the fatigue accumulated through the collective attrition of the other upper body mini studies. However, we are not totally sure why he did not improve his military press even despite this fatigue. These discrepancies underscore the inconclusive nature of our results and the need for a more controlled environment.

Notably, the density progression was adjusted mid-cycle—from 90-second rest intervals to 2-minute intervals—to allow athletes time to complete all four reps at higher intensities. While this improved barbell control and safety–allowing the athletes more rest to increase form proficiency–the change also altered the progression’s stimulus, potentially affecting outcomes.

Limitations

This study’s small sample size (n=4) limits the generalizability of its findings. With only one or two athletes per lift group, individual responses may reflect unique physiological or contextual factors rather than true training effects. For example, a single athlete’s outlier response—whether a large gain or unexpected decline—can disproportionately influence the interpretation of a progression’s effectiveness. A larger cohort would allow for more robust trend analysis and help control for individual variability.

Compounding this limitation, all participants were simultaneously engaged in other high-volume training protocols as part of the Geek Cycle. These included chassis integrity work, a high-frequency pull-up progression, broad jump progression, and a beep test progression. The cumulative training load likely elevated systemic fatigue, reducing recovery capacity and impairing strength adaptation. This overlap makes it difficult to isolate the specific effects of each pressing progression, as performance outcomes may have been influenced more by total workload than by the progression itself.

The study also introduced a mid-cycle change to the pressing density protocol—extending rest intervals from 90 seconds to 2 minutes—after athletes struggled to complete their assigned reps at higher intensities (85–90% 1RM). While this adjustment improved safety and execution, it altered the progression’s intended training stimulus, making it harder to compare early- and late-cycle results or draw consistent conclusions across all participants.

Finally, individual differences in training age, movement efficiency, and recovery ability likely played a role in the divergent results—particularly between Emmett and Seung, both of whom followed the same Bench Press progression. Emmett demonstrated significant strength improvements, while Seung experienced regression in two lifts. This discrepancy may reflect differences in adaptation potential, technique, fatigue management, or baseline familiarity with the trained movement. 

Together, these limitations highlight the need for more tightly controlled study conditions, larger participant groups, and a more isolated training context to draw firmer conclusions about upper-body pressing transferability.

Next Steps

To better determine which upper-body pressing movement offers the greatest strength transfer, this study should be repeated with a larger cohort of athletes and in a more controlled training environment. Future iterations should isolate pressing progressions by removing concurrent training variables—such as pull-up volume, chassis integrity loading, and rock climbing—that may confound recovery and adaptation. Maintaining a consistent progression structure without mid-cycle changes (e.g., adjusting rest intervals) would also improve the reliability of results. Until more conclusive data is available, MTI will avoid labeling any one lift as the definitive “king” of pressing.