Force Plates

A force plate measures two things: force and time. Everything else on that screen is arithmetic. If the arithmetic does not change what you do on Monday, you bought a very expensive scale.

1.The Problem Nobody Warns You About

The force plate is not hard to use. Kids love it. It gives instant feedback, it ranks them against each other, and it makes the weight room feel like a lab. None of that is the problem.

The problem is that it is too generous. A single countermovement jump on a dual force plate produces well over two hundred calculated metrics. No coach on earth can act on two hundred numbers. So one of three things happens, and I have watched all three.

The rule that governs this whole page is the one that governs every other test I run: if the number does not change what happens on Monday, do not collect it. Read the rest with that filter on. I would rather you run four metrics for three years than forty for three weeks.

One honest exception. Sometimes the number changes nothing because nothing is wrong. A flat, healthy board is an answer too. What you cannot do is see a real problem and change nothing, because the kids notice. If nothing changes in the weight room or at practice, a kid learns his jump does not mean anything, and he stops trying on it.

2.What the Plate Actually Measures

A force plate measures force and time. That is it. Every other number on the screen, including jump height, is calculated from those two things. Phil Graham-Smith, who helped build ForceDecks, says it plainly: even jump height is an estimate, whether it comes from flight time or from impulse. Understanding that changes how much you trust a bad trial.

If the athlete is fidgeting while the system weighs him, his mass is wrong, and every impulse calculation downstream is wrong with it. Graham-Smith puts the size of that error at roughly two centimeters of jump height for a 2.2 kilogram body-weight error. That is nearly an inch of fake movement, created by a kid shifting his feet for two seconds. Hold that thought until Test Day.

The four phases of a countermovement jump

Every CMJ has the same four chapters. Knowing which chapter a number comes from is most of the skill.

That four-phase read is exactly how I explain it to people who do not live in this stuff. Load, push, land. Here is the version I gave our nutrition and training staff at Rhode Island in June, word for word off the recording:

Three sentences, three phases, one sport-science staff that suddenly cared. That is the whole pitch. You do not need the jargon to sell this. You need load, push, land.

Output versus strategy

This is the most useful mental split on the page. Graham-Smith frames it as a question of what you asked the athlete to do. You told him to jump as high and as fast as he could, so the honest output metrics are the ones that answer that instruction: jump height, peak power, relative peak power. Everything else on the report is strategy, describing how he produced that outcome.

Two athletes can reach identical takeoff velocity with completely different impulse shapes. Tall and thin, meaning high force over a short concentric time. Or short and wide, meaning less peak force spread over longer. Same height, different athlete, different problem. Strategy metrics separate them, and they are the ones that move first when a kid gets tired.

3.The Metrics That Earn Their Place

Reliability decides what you are allowed to track. A metric that swings six percent week to week when nothing changed cannot tell you about a four percent drop. Before you pick anything, look at what the reliability work actually found.

You do not have to pick one forever. Kids want to know how high they can jump, and that is worth feeding. The split I like: hands on hips on the force plate for readiness, arm swing on the jump mat for the number they brag about. Consistency on one tool, performance on the other.

The five-metric dashboard

One output metric. Three strategy metrics. One safety metric. That is the whole board. Add a sixth only when you can say out loud what decision it changes.

Notice what is not on that list. Not RSI-modified, because of the reliability problem.[4] Not rate of force development, for the same reason. Not the fifty variations of impulse the software offers. You can add mRSI back later as a second readiness axis once you have a year of history and you know its normal swing on your own athletes. Do not start there.

4.Running a Whole Roster Through

Good news first. Throughput is not your bottleneck. VALD's own group testing guide reports thirty players tested in just over twenty minutes on CMJ, and twenty participants assessed in under twenty minutes in a combine-style session. A three-rep CMJ takes fifteen to thirty seconds once the athlete is weighed in. With a clean queue and one set of plates you can move sixty to a hundred athletes an hour. We run our football roster through one plate in two groups in about an hour. The number is real.

Your bottleneck is protocol hygiene. A hundred bad jumps an hour is worse than no jumps at all, because now you have a baseline built on garbage and you will trust it.

Where the jump goes in a full testing day

If the plate is one station among several, test order matters, because fatigue from one test contaminates the next. VALD's guide sequences it like this, and it matches how I run a battery.

1. Warm-up. Standardized across every group.
2. Movement and range of motion. Before anything maximal, because maximal efforts change joint motion.
3. Isometrics. High peak force, very little fatigue cost. Watch for potentiation carrying into the next station.
4. Jumps. High effort, quick, low fatigue accumulation. This is where the plate sits.
5. Sprints and change of direction. Maximal speed, real central nervous system cost.
6. High-fatigue work last. Nordics, repeated sprints, anything that wrecks the posterior chain. Last, always.

Two ways to flow a group

If you stagger, group athletes by size or by testing experience. It cuts changeover time at every station that needs adjustment.

Staffing a station

You do not need five people. You need the roles covered, and on a high school staff one person covers two of them.

One small workflow decision does more for tempo than anything else: reference athletes by number rather than by name at the station. Staff finds the profile instantly and the line keeps moving.

5.Building a Baseline That Holds

A single force plate reading tells you almost nothing. It only becomes information when you compare it to that athlete's own history. Which means the baseline is the most important data you will ever collect, and it is the part everybody rushes.

1. Two familiarization sessions first. Do not record them as baseline. Athletes new to hands-on-hips jumping on plates get better at the test itself for two sessions, and that learning curve will masquerade as a training adaptation.
2. Then three fresh sessions. Fresh means no vigorous activity in the prior 24 hours and no hard training in the prior 48 to 72. Three max reps each, hands on hips, self-selected depth.
3. Average them. Both within a session and across the three. A baseline off one hot Monday makes every athlete look tired for the rest of the year.
4. Record the standard deviation too. The spread is what lets you set a real flag later instead of borrowing someone else's percentage.
5. Let it drift. Roll the baseline on a 30-day window in-season rather than freezing August forever. Athletes change. A baseline that never updates turns into a grudge.

One more thing about timing. Do not test during camp. Everybody is cooked, the numbers mean nothing, and you will spend the season comparing against a floor. Test going into week one instead, where a good number confirms the taper worked.

And do not start the plates cold in the middle of a season. If you missed the window, do not drop a brand-new test on your kids in week three. It is a new routine they have not learned, and the first few sessions will be learning, not data. Wait for the bye week or the off-season, start clean, and build the baseline properly. That is exactly where we are this year.

6.Reading Fatigue

A fatigued athlete does not immediately jump lower. He compensates. He dips deeper and spends longer on the ground building the same impulse. The sheet reads normal. The kid is a step slow on Friday. That is not a theory, it is what the force-time curve shows, and the research backs the direction.

That is the frame I want you to coach from:

The firing order

Read your board in this order. Not the order the software prints it.

Same read, in plain terms: 22 inches in half a second is ready. 22 inches in nine tenths of a second is a kid working twice as hard to stand still. Read the height alone and you pass the second one.

Setting the flag

You have two honest options. Pick one and stay with it.

Either way, act on two consecutive flagged readings, not one. Normal day-to-day swing is 5 to 8 percent. One bad reading is a bad night of sleep or a fight with a girlfriend. Two in a row is fatigue.

And read the individuals, not the mean. A camp study logged a group mean drop of 1.5 percent on a day when individual athletes ranged from minus 27 percent to plus 8 percent. The team average said everything was fine and missed both ends of the roster. My standing instruction to my own staff is to count the athletes who cannot produce a valid jump at all, not just to read the ones who can.

One honest caveat about the word fatigue

Dan Cohen makes a point worth sitting with. Practitioners overuse the term fatigue monitoring, and it primes you to go looking for bad news. The Premier League data he worked on showed elite players getting better neuromuscularly across a competitive season when they were managed properly. His preferred framing is load response, not fatigue.

That matters for how you talk to your head coach. If every report you hand him is a list of problems, he stops reading them. Show the guys who are climbing too.

It matters even more for how you talk to the kids:

Down is not the same as broken. Tell a kid he is a little down today and here is what we are changing. Do not tell him he is fatigued and let him decide what that means.

7.Asymmetry, Honestly

This is the section that will cost me something, because I run the 10 and 15 percent thresholds on my own board at Rhode Island. Here is what the evidence actually says about them.

Those bands are the industry convention. Hawkin Dynamics uses them, NordBord reporting uses them, my app uses them. They are useful, consistent, and easy to teach. They are not a validated injury threshold, and anybody who sells them to you as one is overselling.

So what do you actually do

You keep measuring it. You stop treating the number as a verdict. Four rules make asymmetry useful again.

Here is how that becomes a coaching decision instead of a spreadsheet entry, in my own words from the same staff meeting:

That last sentence is the entire value proposition of the plate. Without it, everybody gets the same RDLs. It is not that the RDLs are wrong. It is that you are guessing, and guessing is what we are trying to stop doing.

The football-specific requirement

Normalize to body mass. Always. A 310 pound offensive lineman and a 180 pound corner cannot be compared on raw Newtons, and any leaderboard that does it will hand your line every top spot and tell you nothing. Force and impulse metrics go to per-kilogram before they go on a board that crosses position groups. This is more important in football than in almost any other sport, purely because of the size spread on one roster.

It cuts the other way too, and this is the conversation you will have every spring:

Jump height alone would tell that kid he wasted a year. The plate tells him he got a lot stronger. That is a retention conversation, not just a data point.

One more note on big bodies. Heavy athletes often show a deeper, longer braking phase, and that is frequently a smart adaptation rather than a fault. They are avoiding a force spike their joints have to eat. Do not read a lineman's long eccentric duration through a defensive back's standards.

8.Profiling: DSI, EUR and RSI

Monitoring asks how is he today. Profiling asks what kind of athlete is he and what should his block look like. Different question, different cadence. Profile at milestones: start of preseason, end of preseason, end of a training block. Do not run these weekly.

Dynamic Strength Index

DSI divides CMJ peak force by isometric peak force from a mid-thigh pull. It answers one question you cannot answer by eye: is this kid rate-limited or force-limited.

Now the caveat, because it is a big one. Suchomel tested 88 male and 67 female NCAA Division 1 athletes and found DSI correlates very largely and negatively with IMTP peak force (r = -0.848 in men, -0.746 in women) but only small to moderately with CMJ peak force (r = 0.297 and 0.313) and trivially with CMJ peak power (r = 0.008 and 0.191).[10] In other words, DSI is mostly a readout of the isometric denominator. A low DSI usually means the athlete pulled a big IMTP, not that his jump was bad.

Their case-study analysis made the same point from the other direction: athletes with near-identical DSI scores had completely different percentile profiles and different force-time curves, and needed different training. So use DSI as one input, never as a sorting hat. Pair it with reactive strength and a 10-yard split before you assign a block.

For the record, I do not run DSI. We do not have a good mid-thigh pull setup, and a kid who is great at the isometric pull ends up with a DSI that looks bad when nothing is wrong with his jump. If you have only the plate, use EUR below instead.

Eccentric Utilization Ratio

EUR compares a countermovement jump against a static squat jump, where the athlete holds the bottom position for a five count before jumping. The difference between them is what the stretch-shortening cycle is contributing.

• Around 10 percent higher on the CMJ is typical.
• 20 percent or more means heavy reliance on elastic energy with less pure concentric force underneath. Build the strength base.
• Near zero difference means he is not using elastic energy at all. Reactive work, plyometrics, fast stretch-shortening drills.

EUR's real virtue is that it needs no isometric rig. Two jumps on the same plate, and you get the profiling question answered without owning an IMTP setup.

Reactive Strength Index

RSI is jump height divided by ground contact time, measured off a drop jump. It scores how much output the athlete gets for the time he spends on the ground, which is the quality football actually runs on. Two kids with the same 30 inch vertical, one off a stiff fast contact and one off a long soft absorb, are not the same football player.

Two cautions. First, RSI is only comparable against the same drop height, so record the box height and hold it. Second, remember that modified RSI from a countermovement jump is a different, noisier metric with poor test-retest reliability in the force-plate comparison work.[4] Do not treat RSI and mRSI as interchangeable.

If you add reactive strength, my preference is the old-fashioned version: a drop jump off the same box every time, stick the landing, read ground contact time and height. Modified RSI is easier because it is just a countermovement jump, but a slow start off the ground skews it and it is the noisier metric. Add it after a year of clean jump data, once the kids know the plates. You do not need it on day one.

9.The Board, and How to Fund It

Here is what my actual coach-facing view looks like at Rhode Island, so you can copy the shape rather than invent one.

What the board caught

Two reads from our own sorting, to show you why the sort buttons matter.

On the trends view, the goal for asymmetry is a flat line. One of our guys went down to 2.4 percent over three tests. That is what getting better looks like on that chart.

Where we honestly are

I want you to know the real state of this, so you do not think you are behind. We had old Hawkin plates that did not work well. This summer we got one VALD plate on a trial, and it is our first real year. We tested the roster through the summer, then the school network problem kept us off the plates once camp started, and I will not drop a new test on the kids in week three. We retest on the bye week.

We still use jump mats every week. We have used them since I took over in 2021, the kids know the routine, and that history is worth keeping. This year the plate is data collection. Next on our board: countermovement depth and time to takeoff, so the trends view covers the full firing order instead of just height and asymmetry. By this time next year we will have the process locked in. You can start at the same point we did. Start testing, teach the kids what the numbers mean, and change the program when the numbers tell you to.

You also do not need a sports science department to build a board like mine. VALD has its own dashboard, and it is fine. I built ours with Claude, with some help from ChatGPT. Export the numbers and tell it what you want: "You have the numbers. Build me a report. Build me a dashboard that tracks how these kids trend over time." That is how my board got built.

The report is the product, not the plate

I brought GPS data to our building in 2022 and it fell on deaf ears. The conversations did not turn until 2024. Two years. The tool did not buy me credibility. The report did, and specifically the report that told somebody what to do differently this week.

So build the loop before you build the dashboard. Data goes to the athletic trainer with a name and an assignment attached. Then you go back and ask what changed. Here is that exact conversation at our place:

That is not a shot at an athletic trainer. That is what new information is supposed to do to everybody in the building, including me. If nobody's job got harder when the plates showed up, nobody is using them.

The AD pitch

Do not walk in with metrics. Walk in with load, push, land, then a number, then a cost. Our real numbers, so you have a benchmark: $2,800 a year for hamstring testing, $3,500 a year for the jump plates, $6,300 a year for both. Ours arrived as a summer loaner, and the deadline was the pitch. A free trial with an expiration date is the single best lever you will ever get for this conversation, so if a vendor offers one, take it and put the return date in front of your AD early.

Then do the math out loud, because it is not close. One missed game from one preventable non-contact soft-tissue injury costs a program more than $6,300. Not in medical bills. In a starter who is not available, in a season that turns on one Friday, in a kid whose senior year gets taken from him. That is the argument. Not the metrics.

If the answer is no, the answer is no for now. Go run the no-budget version below, build the report anyway, and ask again in a year with a season of your own data attached. We go to war with the army we have.

§Objections and Questions

§Glossary

The terms used on this page, in the order you will meet them on a ForceDecks report.

Countermovement jump (CMJ)

A vertical jump that starts from standing, dips, and rebounds. The dip is the countermovement. The standard monitoring test.

Squat jump (SJ)

A jump from a held bottom position with no dip. Removes the stretch-shortening cycle so you measure pure concentric force.

Ground reaction force

The force the ground pushes back with. The only thing a force plate directly measures, alongside time.

Impulse

Force multiplied by the time it is applied, in Newton-seconds. Net concentric impulse decides takeoff velocity, which decides jump height.

Eccentric phase

The lowering and braking portion of the jump. Muscles lengthening under load.

Concentric phase

The propulsive portion from the bottom to takeoff. Muscles shortening under load.

Countermovement depth

How far the center of mass drops during the dip. A fatigued athlete usually sinks deeper.

Force at zero velocity

Force produced at the exact bottom of the dip, where downward motion stops. A sensitive fatigue marker.[2]

Rate of force development (RFD)

How quickly force rises, in Newtons per second. Conceptually vital, but poorly reliable between sessions and between systems.[4]

Reactive Strength Index (RSI)

Jump height divided by ground contact time, from a drop jump. Only comparable at the same drop height.

Modified RSI (mRSI)

Jump height divided by time to takeoff, from a countermovement jump. Safer than a drop jump, but noisier than RSI and mixes two unlike quantities.

Dynamic Strength Index (DSI)

CMJ peak force divided by isometric peak force. Largely driven by the isometric denominator.[10]

Eccentric Utilization Ratio (EUR)

Countermovement jump compared against squat jump. The no-isometric-rig substitute for DSI.

Isometric mid-thigh pull (IMTP)

A maximal pull against a fixed bar. Gives an absolute strength number without putting a novice under a heavy barbell.

Coefficient of variation (CV)

How much a measurement bounces around when nothing has changed, as a percentage. A metric cannot detect a change smaller than its own CV.

Smallest worthwhile change (SWC)

The smallest change that matters in practice. A metric is trustworthy when its typical error is smaller than its SWC.[3]

Intraclass correlation coefficient (ICC)

How consistently a test ranks the same athletes across repeat sessions. Above 0.700 is the usual working floor.

Asymmetry percentage

The difference between limbs, expressed against the stronger side. Sign convention on my board: negative is left dominant, positive is right dominant.

Deeper Reading

Practitioner and vendor sources used on this page. Named separately from the peer-reviewed references below on purpose. These are experienced people and useful material, but they are not research, and the numbers in them have not been through review.

References

Talk About It Inside