You lift more than your partner. You have the better log. Then you both throw for the same crimp off the same deadpoint, and they stick it and you don't, three times in a row.
Nothing is wrong with your testing. You're measuring one quality and the move is asking for a different one.
Force is not instant
When you pull on something as hard as you can, your force doesn't arrive all at once. It builds. Give a finger a couple of seconds of maximal effort and it reaches a peak, which is exactly what a max lift measures and why our testing protocol uses a three-second hold: peak force lands early and everything after that is fatigue resistance in a strength test's clothing.
A deadpoint doesn't offer you a couple of seconds. Your hand arrives on the hold, and the window in which you either hold it or don't is a fraction of one. How much force you can produce in that window is a separate quality from how much force you can reach eventually, and it has a name: rate of force development.
Why the two come apart
Two climbers can reach the same peak. If one of them gets to eighty percent of it noticeably faster, they will hold latches the other one drops, while the log says they are equally strong. The log isn't lying. It's answering the question it was asked.
This is not a hypothetical distinction. Our post on CARC-style endurance training documented a self-experiment where four weeks of low-intensity endurance work left the trained arm about 125 milliseconds slower to reach 80 percent of max strength, with max strength itself slightly up. One quality moved in one direction, the other moved in the other, on the same hand, in the same month. Speed of force and size of force are trainable separately and they can drift apart without you noticing, because most people only measure one of them.
Why more max work doesn't fix it
The instinct when you drop latches is to go get stronger. It isn't wrong, exactly, and a bigger ceiling is rarely a problem. But heavy, slow lifting trains you to grind up to a high number. Latching trains you to arrive there immediately. If every finger session you run is a slow build to a heavy hold, you have spent months practising the wrong half of the skill.
The thing that changes it is intent: pulling as hard and as fast as you can from the first instant, rather than easing into the load. Our recruitment pulls post owns that method in full, including why a deeper edge exists specifically to let you pull with maximum intent without the edge itself becoming the limiter.
The same idea shows up in our overcoming versus yielding isometrics post: how you apply force to an immovable object is itself a training variable, not a detail.
The measurement problem
You can't see rate of force development with plates. A plate stack tells you whether the load left the floor. It has no opinion about how quickly you got there, which is the entire quantity in question. A load cell that logs a force trace can show you the slope of your pull, and our gauge post treats that as one of the genuine reasons to own one.
Not owning one changes less than you think. You can train the fast end without measuring it, the same way you can train endurance without a lab. What you lose is the ability to prove it moved, so you fall back on the climbing: are you sticking latches you used to drop? That's a slower signal and a noisier one, but it is the outcome you actually wanted.
One caution before you go and pull fast at your limit
Maximum-intent pulling is the highest-stress thing you can do to a finger, and it stops being a good idea when there is no base underneath it. Our pulley post covers why load without a base to absorb it is exposure rather than training, and the timelines post covers why your nervous system will happily drive your fingers harder than your tissue is ready for. If you're new to structured finger training, build the base first with something like our 8-week no-hang program and come back to fast pulls later.
Warm up first and stop on pain. Full safety guidance.
The edge that lets you pull that way
Pulling with maximum intent needs an edge that isn't itself the limiting factor. If the edge is small enough that your skin and your grip give out before your recruitment does, you're measuring the edge rather than yourself, and the whole point of the exercise is lost.
Our Dual Edge 28MAX Recruitment Edge is a 28mm loading edge built for maximum force production. The deep, flat profile minimizes grip limitations, making it ideal for max hangs, heavy pulls, and recruitment work. Here that depth is doing a diagnostic job rather than a training one: it is what lets you find out whether your problem is the size of your force or the speed of it, without the hold confounding the answer. How to run the pulls is in the recruitment pulls post.
The bottom line
Your max lift measures how much force you can reach. Your latch measures how fast you get there. They are different qualities, they are trained differently, and only one of them shows up in most people's logs. If you are strong on the block and late on the hold, the missing work is fast, maximally intended pulling, not another block of grinding heavy holds. Build the base first, then train the speed on purpose, and stop treating a dropped deadpoint as evidence that you need a bigger number.