The highest-value fix on this list is not pulling more often — it's ending every set one or two reps short. In Pareja-Blanco's 8-week trial, the group that cut every set at 20% velocity loss gained the same strength as the group that ground on to 40% velocity loss, on about 40% fewer reps. Frequency, which I led my own reel with, is a delivery system for volume rather than a lever by itself — and for a posterior-chain lift, the volume-equated data is null. So the prescription is short: two to three hinge exposures a week (only one needs to be a heavy conventional pull), every work set stopped at 1-2 reps in reserve, and weight added only when every set hits the top of its rep range.
I made a reel with these eight tips. This is the long version with the studies attached, in a new order — because three of my eight tips don't have a trial behind them, and I'll tell you which three. And if you're chasing a specific number rather than a bigger one, here's where a 315 deadlift actually ranks before you start adding plates.
Why is my deadlift not going up?
Because the last rep of every set is a grind, and that rep comes out of the sets after it. This was tip #2 in the reel. It should have been tip #1.
Pareja-Blanco 2017 ran 22 young men through eight weeks of velocity-based training, one group stopping each set at 20% velocity loss, the other at 40%. Same strength gains, about 40% fewer reps in the 20% group, plus a bigger countermovement jump improvement (9.5% vs 3.5%, significant at P < 0.05). Two honest caveats: that was a squat protocol, not a deadlift one, and the 40% group got more vastus lateralis hypertrophy. Extra reps past that point buy size, not strength.
Davies 2016 pooled the failure literature and found an effect size of 0.34 favouring stopping short (p=0.02) — while calling that difference "unlikely to be meaningful" in the authors' own words. Read it precisely: failure is unnecessary, not harmful. Refalo 2024 shows where the cost lands: 18 trained lifters did eight weeks of leg press and leg extension with one limb trained to failure and the other stopped at 1-2 RIR, and velocity loss and rep loss inside the session were consistently greater on the failure limb, though eight-week total volume ended up similar. The barbell version of that measurement exists too, and it's blunter: six bench sets at 75% of 1RM cost 7.58 reps from the first set to the last when they went to failure, versus 2.79 at 3 RIR — with total reps landing in the same place either way, which is the whole argument for how to increase your bench press without maxing out every session.
"Train it like a skill, not a test" survives the fact-check intact. Here's where to stop on a set in more detail.
How often should you deadlift?
Two to three hinge sessions a week — but my reel said "once a week is why you're stuck," and for a posterior-chain lift, that line isn't supported.
Grgic 2018 pooled 22 studies and found strength effect sizes climbing with frequency: 0.74 at once a week, 0.82 at twice, 0.93 at three times, 1.08 at four or more (p=0.003). Then the same paper takes it back twice. The volume-equated subgroup showed no significant effect (p=0.421), and the lower-body subgroup never reached significance (p=0.070) — legs and hips being the exact region a deadlift lives in. The upper-body subgroup did clear significance (p=0.004), which is why the same argument lands differently when the lift is the overhead press.
The direct tests agree. Johnsen and van den Tillaar 2021 split the same weekly volume across two versus four sessions in 21 trained men and found no frequency effect on squat 1RM (p=0.640). Colquhoun 2018 put 28 trained men through six weeks of equated powerlifting training at 3 versus 6 sessions a week: deadlift 1RM rose +19 kg versus +21 kg, with no significant difference between groups. Double the sessions, same result.
So keep 2-3x for the honest reason — more exposures mean more room for quality hinge volume without one session becoming a two-hour grind. And here's the part nobody writes: only one of those needs to be a heavy conventional pull from the floor. RDLs, block pulls and hinge variations count as exposures. The same logic runs across every muscle group, not just the deadlift, and four training days is plenty to hold it. If you want two or three hinge days inside a week you'll actually run, that's what the Momentum Method layout is built around.
Why is my deadlift weaker than my squat?
Usually because of how you spread your volume, not how you're built. Most people squat two or three times a week and pull once, then wonder why one number moves.
Strength does travel between the two. In Plotkin and Contreras 2023, nine weeks of back squats or hip thrusts improved deadlift 3RM about equally — +15% versus +16% — with similar gluteal growth in both groups. Read it as a direction rather than a number to expect, because the participants were untrained. But if squatting carries over to the deadlift at all, a lagging pull isn't an anatomy verdict. It's a practice gap: fewer reps of that specific skill, at that specific bar height, than of the lift you do three times a week. The fix is the prescription above — more hinge exposures, each ended before it turns into a grind, logged so you can see whether the last four weeks moved anything.
Does your deadlift setup actually add pounds to the bar?
It makes the rep repeatable, which is not the same claim as adding weight — no trial shows a setup cue adds pounds to a max.
Edington 2018 tested maximal isometric conventional deadlifts across different setup positions and found differences in EMG and joint kinematics but no difference in total ground reaction force, or in shear and compression at the joints. Position changed what the muscles were doing. It didn't change how much force came out the bottom.
What setup buys you is a position that survives load. A 2025 Frontiers in Sports and Active Living analysis found lumbar kyphosis increases as the bar gets heavier, tracking with posterior pelvic tilt — you round more at 90% than at 60%, whether or not you notice. That's the real argument for setup rigor: position degrades as load rises, so start from the strictest version you have.
"Take the slack out" and "push the floor away" are cueing language, not findings. Keeping the bar close to your shins is neither — that one is physics, a shorter moment arm at the hip.
Do deficit deadlifts increase your deadlift?
This is the weakest tip of the eight, and I called it "huge" in the reel. No randomized trial shows that deficit deadlifts raise a conventional 1RM. The source most articles lean on is a 2019 Strength and Conditioning Journal technique piece — a coaching article on how to perform the lift, not an intervention study.
The mechanism is reasonable, and range of motion clearly changes what you can lift: a 2023 JSCR technical note in 18 college wrestlers measured partial-range deadlifts from just above the knee at 226 kg against 192 kg from the floor, correlated at r=0.85. But that's a measurement, not a training result — the authors' point was that you can predict one from the other. Reasonable is not the same as demonstrated.
I like deficit pulls, and the 1-2 inch plate in the reel was my own pick, not a researched dose. But I can't point you to a study where deficit deadlifts raised anyone's conventional max, so I'm not going to sell them to you like I can.
Do your lats really matter on a deadlift?
The cue survives. The superlative doesn't — I called a loose upper back "the number 1 reason deadlifts break down," and there's no ranking anywhere that supports that.
Look at what the lat does. It contributes little to lumbar extension and nothing at all to hip extension, the movement that finishes a deadlift. Its job on a pull is isometric: hold the bar against your body and stop your upper back from folding. That makes it a position cue, not a strength cue.
Both studies above back that reading. In Edington 2018, the differences between setups showed up in muscle activity and not in force output. And the 2025 Frontiers analysis shows rounding increasing with load — the exact thing the lats are there to resist.
So keep the cue exactly as I said it: squeeze your lats like you're crushing something in your armpits. Then go build a bigger back so there's something there to squeeze.
What accessory lifts actually carry over to the deadlift?
This is the best-supported accessory tip of the eight. Hip thrusts transfer to the deadlift about as well as squats do.
In Plotkin and Contreras 2023, nine weeks of hip thrusts raised deadlift 3RM +16% against +15% for back squats, with similar gluteal growth in both groups. The participants were untrained, so read the direction, not the number.
Then the twist. A 2025 blinded JSCR trial put 36 untrained people through nine weeks of Nordic curls or stiff-leg deadlifts and found hamstring hypertrophy played a minimal role in how strength transferred between the two. Bigger hamstrings did not automatically mean a stronger pull. What accessories buy you is pattern strength — practice of the hip hinge under load — not size that converts on its own. Worth knowing while you build bigger legs, because the size and the transfer are separate purchases.
Honest note on my own list: RDLs and hip thrusts have transfer evidence. Back extensions don't. No trial has tested them against a deadlift max.
Should you lower the bar or drop it?
This one is preference, and I wrote it in the reel like it was evidence. It isn't.
Schoenfeld 2015 pooled the tempo literature in Sports Medicine and found hypertrophy was similar across rep durations from 0.5 to 8 seconds per rep. Slower is not automatically better. And the specific question — touch-and-go versus dead-stop deadlifts — has never been run as a training trial. Nobody knows.
There is a real trade-off, though, and it connects straight back to the first fix: a controlled eccentric adds eccentric fatigue, and eccentric fatigue is a cost you pay in the next session. If you're pulling two or three times a week, that matters.
My reasons for lowering the bar are mine — I like arriving in the same position every rep, and I don't want to smash the platform. Those are fine reasons. They're just not findings.
Is grip strength what's limiting your deadlift?
Half right. Grip caps how much deadlift volume you can accumulate; no trial shows grip work raises a max.
The strap studies make the volume case cleanly. Jukic 2021 found that in 16 trained men, straps improved barbell velocity, grip retention, RPE and grip security in the deadlift. A 2023 JSCR trial in trained women found more reps completed with straps and grip preserved, with no cost to barbell velocity. What those prove is that your hands end your sets early — not that farmer carries add pounds to a one-rep max. Nobody has tested that.
That splits the fix in two. If the problem is volume — sets ending because the bar rolls out of your fingers — straps are the fast fix, and using them on back-off sets costs you nothing measurable. If the problem is a competition where straps are banned, grip work is the slow fix: double-overhand holds, farmer carries, hex bar pulls. Pick the one that matches the problem you have.
The eight, reordered
The same eight, reordered, with the evidence attached:
- Stop 1-2 reps short of failure — backed (Pareja-Blanco, Davies, Refalo).
- Two to three hinge exposures a week, one heavy — partly backed; frequency buys volume, nothing more.
- Build the posterior chain outside deadlifts — backed for hip thrusts and RDLs, untested for back extensions.
- Squeeze the lats — partly backed; a position cue, not a strength cue.
- Fix your setup before you pull — partly backed; it makes the rep repeatable as load degrades position.
- Train grip — partly backed; straps solve volume, grip work solves the competition problem.
- Deficit deadlifts — preference; plausible mechanism, no trial on a conventional max.
- Control the descent — preference; my reason, not a finding.