The question

A handcuff is the backup who would inherit the job if the starter in front of him went down. The usual pitch is insurance. You spent a second-round pick on a running back, so spend a thirteenth-round pick protecting him. We’ve always advised against this, but now we have the data.

We ran it. We looked at nine regular seasons from 2015 through 2018 and 2020 through 2024, covering every team-season where a clear lead back emerged in the first four weeks and a clear primary backup sat behind him. That gave us 283 team-seasons. For each one, we looked at what the backup actually returned and how much the starter’s health changed the answer.

What the backup returns

For comparison, over the same seasons any bench-caliber back who was not a designated handcuff finished RB24+ 10.5% of the time and RB36+ 21.4% of the time. So the handcuff gives you roughly three extra percentage points of hit rate over a random late-round back.

That is a slight edge, but it is much smaller than what most people think they are buying.

The bet only cashes in the extreme case

Handcuffing feels smart because lead backs do miss time. Only 30% of them played every game after Week 4.

But missing one or two games does almost nothing for you over the course of a season. The backup’s season-long outcome in that group is basically the same as it is when the starter stays healthy, because a two-week fill-in is not enough to move a full-season finish.

What you really need is for your starter to miss six or more games, and that happened in only 15.2% of team-seasons. Even then, the backup finished as a top-24 back just 23.3% of the time.

Put those two numbers together and the pure insurance case gives you a startable RB2 in roughly 3 or 4 out of every 100 handcuff picks.

What it pays when it hits

Now take the cases where the lead back missed three or more games and look only at the weeks he was out. The backup averaged a median of 10.1 points per game against an RB36 flex baseline of 7.7. He beat the flex baseline in 67.3% of those situations and reached true RB1 weekly production, 17-plus points per game, in 11.2%.

That is the important part.

The handcuff often works the way you expect when the opportunity actually arrives. The problem is that the payoff is usually a competent flex starter for a few weeks, not a league-winner, and you had to hold a bench spot all season waiting for that specific event to happen.

You are protecting yourself from a bad outcome, not creating much upside if everything goes right.

The result that flips the usual advice

Now split the group where the starter missed three or more games by how much work the backup already had in Weeks 1 through 4.

The cleaner the handcuff, the worse the results.

The backup behind an unquestioned bell cow finished RB24 or better in only 6.5% of cases, even when the bell cow missed a third of the season. The back who was already splitting the backfield hit at 29.3%.

That makes sense once you look at what each player was doing before the injury. The best predictor of a late-round back’s season was touches he already had, not the injury he was waiting for.

The pure backup gets his opportunity only after something goes wrong and often fails to hold the job, sometimes even losing work to a veteran signed off the street. The committee back is already producing in September and simply gets more work if the starter goes down.

So what do you do with the pick

• Do not draft your own starter’s backup just for protection. You already own the risk tied to that backfield. Adding the backup spends another roster spot on the same offense and leaves your best-case outcome as “my season is not ruined.”
• Draft backfield ambiguity instead. A back who already owns a third or more of a real backfield can cost the same late-round pick and hit the RB24 line about four times as often.
• If you want the injury lottery ticket, buy someone else’s. A backup you own without the starter costs you only a bench spot when the starter stays healthy, but becomes a tradeable asset if that starter goes down. Handcuffing your own back ties two roster spots to the same offense.
• Compare the handcuff against the bench receiver you are giving up. Bench-caliber receivers hit WR36+ at 13.4%, so a good handcuff can still clear that bar. This is not an argument for never rostering a backup. It is an argument for ranking him honestly instead of giving him a pass because he is “insurance.”

The bottom line

Be suspicious anytime someone calls handcuffing non-negotiable.

The data says it can prevent a total collapse at running back in a small percentage of cases, and when it works, the payoff is usually a flex-caliber player for a few weeks. The version people recommend most often, the clean backup behind a workhorse, is also the version that performed the worst.

If you want the late-round back who can actually change your season, stop asking who is next in line and start asking who already has the ball.