How strong should a runner be? Standards by bodyweight

No study has found a strength level a runner needs to reach. The trials that made runners more economical didn't set a target. They took runners from wherever they started and made them stronger with heavy, low-rep work: half squats at a 4-rep max, three times a week for eight weeks, raised 1RM by 33% and running economy by 5%.¹ If you want a number to aim for anyway, a squat of about 1.5 times bodyweight for men and 1.1 times for women is a fair working target. That's hybridAF's intermediate standard, and it's our cut point, not a finding from running research.

What the running studies actually asked for#

The best-known trial is Støren and colleagues. Eight well-trained runners, four men and four women, added four sets of four-rep-max half squats three times a week to their normal running. After eight weeks their 1RM was up 33.2% and their running economy 5.0%. Time to exhaustion at maximal aerobic speed rose 21.3%. VO2max and body weight didn't change, and the running-only controls didn't change at all.¹

That pattern holds across the wider literature, with less consistency. Blagrove's review found 24 studies that met its criteria. Running economy was measured in 20, and it generally improved by 2 to 8% against controls, though not in every study. Time trials from 1.5 to 10 km tended to improve. VO2max and body composition were typically unaffected, and the authors suggest two to three strength sessions a week.²

The more recent meta-analysis from Llanos-Lagos split the methods apart. High-load training, defined as 80% of 1RM or more, gave a small improvement in economy (effect size -0.266). Combining methods gave a moderate one (-0.426). Submaximal loads of 40 to 79% didn't improve economy, and neither did isometric work. The high-load effect was bigger at faster speeds and in runners with a higher VO2max.³

Notice what all of these prescribe. It's a percentage of your own max, or a rep max. None of them says a runner should squat a certain multiple of bodyweight. The answer the research gives to "how strong" is "stronger than you are now, using heavy loads", which is useful but doesn't hand you a target.

Strength standards by bodyweight#

So the target has to come from somewhere else: what strong people of your sex and size lift. These are the cut points hybridAF grades against. Each is your estimated one-rep max divided by bodyweight.

Each level maps to a point on a 0 to 100 ladder: untrained 20, novice 40, intermediate 60, advanced 80, elite 95. A lift between two levels gets a score in between. Your Strength Index is the plain average of the lifts you've logged, so one lift is enough to get a number and more lifts make it fairer.

Squat standards, novice to advanced

	Low (× bodyweight)	High (× bodyweight)	Mark (× bodyweight)
Men	1	2	1.5
Women	0.7	1.5	1.1

hybridAF's squat cut points (src/lib/score-grading.ts, STRENGTH_STANDARDS). The bar runs from novice to advanced; the mark is intermediate. Our design choice.

Read the mark as the working target. For a 70 kg man, intermediate is a 105 kg squat, and the novice-to-advanced band runs from 70 to 140 kg. Plenty of runners who have never lifted will start below the bar, which is fine. The economy trials measured gains from wherever people started.

If you haven't entered a bodyweight, the same levels are graded on absolute kilos instead, which ignores your size. For men, intermediate is then a 100 kg squat, 125 kg deadlift, 70 kg bench and 48 kg press; for women, 55, 70, 32 and 26 kg. That's cruder, so add a weigh-in if you want the ratio.

There's one more thing the table doesn't do. Strength isn't adjusted for age. The engine half of the score is graded against age-specific VO2max norms; the strength half isn't, so a 50-year-old is held to the same ratios as a 25-year-old.

How those numbers compare with powerlifters#

The most useful reference we have is van den Hoek's normative data: 809,986 entries from drug-tested, unequipped powerlifting meets. Relative strength peaked between 18 and 35.⁴ The abstract gives the 90th percentile for that age group, and here they are next to our elite cut points.

So our elite level sits a little below the top tenth of competitive powerlifters for squat and deadlift, and roughly level with them for the men's bench. That's deliberate. These are people who train for nothing but those three lifts, and "elite" for someone who also runs shouldn't need a powerlifter's whole career. Turning those norms into five rounded cut points is our judgment, and the science page lists it as a design choice.

There's also a fair criticism of dividing by bodyweight at all. Ratio scaling slightly favours lighter lifters, because strength doesn't rise in proportion to size. We chose it because a ratio is something you can check in your head.

Why women get different numbers#

The women's ratios are lower, and for the lower-body lifts, body composition explains most of it. Bartolomei compared 16 men and 14 women who all trained strength seriously. Once their 1RMs were adjusted for lean body mass, there was no significant sex difference in the squat or the deadlift. The bench press still differed, with men stronger even per kilo of lean mass.⁵

That fits the table. The women's squat and deadlift standards sit closer to the men's than the bench does. It's a small study, though, and it explains the gap rather than setting the cut points.

What counts, and what doesn't#

The standards are for free barbell lifts, so that's all the score grades. Dumbbell, machine, Smith machine, cable and kettlebell versions don't count, and neither do leg press, hack squat, goblet squats or one-leg work like split squats and Bulgarians. Front, box and pause squats do count: they're barbell squats, and most people lift less on them, so they can only understate you.

The trap bar is the one people argue with. Most people lift more with a trap or hex bar than with a straight bar, so grading it against barbell deadlift numbers would flatter the result. It stays out. That's a statement about the yardstick, not the lift. If your program is trap bar deadlifts and split squats, keep doing it. You just won't get a strength grade from it until you log a barbell squat, bench, deadlift or press.

There's also a sanity ceiling. A logged lift above 3.5 times bodyweight on the squat, 4 on the deadlift, 2.6 on the bench or 1.5 on the press is treated as a typo and dropped. Those ceilings are well above the elite cut points, so they only catch mistakes, like a weight logged in pounds as kilos.

The estimate itself comes from your best set of 1 to 12 reps through the Epley formula. How far to trust that estimate is a separate question, covered in Epley vs Brzycki.

My own lifts against the table#

My Strength Index is 74. It's the average of three graded lifts: bench 90, deadlift 78 and squat 53. No overhead press counts: the one press in my log is far above the sanity ceiling, a logging mistake, so the score drops it. My bench is advanced to elite, the deadlift is just short of advanced, and the squat sits between novice and intermediate.

So for a runner, the lift the research leaned on is my worst one. Støren's trial was built on the half squat, and the squat is my lowest grade. My squat grade also comes from a set I logged in August 2025. The score keeps your best logged lift with no time limit, so it won't fall if my squat has faded since. That's a reason to retest it, not a reason to trust it.

How strong is strong enough if you also run#

This part is our opinion, because the research doesn't answer it.

Intermediate on the squat and deadlift is a sensible floor for a hybrid athlete, and it's the same bar the definition of a hybrid athlete we use sets for both halves. Past that, more strength keeps helping the strength half of your score, but whether it helps your running is unknown. No trial has compared runners at 1.5 times bodyweight with runners at 2.0.

What the research does support is the dose: two or three sessions a week, at or above 80% of your max, kept up for weeks rather than days. The programmes in the meta-analysis ran from 6 to 24 weeks.²³ Does lifting make you a faster runner? goes through those trials in detail. If you want to see where your lifts land right now, the strength standards by bodyweight grade them against the same table as above. Whether your lifting or your running is the bigger limit is a different question, answered in strength-limited or engine-limited.