Optical timing gates are precise, accurate instruments. The design and manufacture of the technology is wonderful. They are simple, portable and reliable. They display results immediately. Researchers use them. Coaches use them. Teams, leagues and associations use them. But not everyone is using them as they should, and therein lies the problem with optical timing gates.
More often than not, users fail to measure what they intend to measure with optical timing gates. This leads to faulty conclusions, which can result in bad training plans, and maybe even the selection of slower athletes.
This article will start with a brief overview of how they work, which will lead into what they can and cannot measure. Next, how good research led to bad, more bad, and then worse. Finally a number of fixes so you can actually measure what matters —the time it takes for a player to move from where they are to a target destination, also known as: Speed.
Very simply, a light beam is sent from a laser or infrared LED horizontally across a gap of a metre or so to an optical eye sensor. When that beam of light is interrupted, the sensor sends an electrical signal via wire or radio wave to a digital timer. Typically one sensor starts the timer, and a second timer stops it. If the distance between beam-sensor 1 and beam-sensor 2 is known, then it can be divided by the time elapsed between the two signals to calculate average speed.
Elapsed time and average speed over a given distance. That is what two gates can measure. That is all they can measure. Not actual speed, but a calculated average speed. They do not tell you if the moving object or person was accelerating, coasting or braking. Optical timing gates do not tell you if the athlete moved in a straight line between them or deviated slightly, or left the track to get an ice cream and came back 10 minutes later. They only let you know the time elapsed between the triggering of beam-sensor 1 and beam-sensor 2.1
What’s the big deal? Any kid who leaves in the middle of testing to get a snack has not place on my team either.
If you set up a bunch of gates in series, you can get split times every 5m then use them to calculate approximate accelerations and velocities across various increments of the total displacement. You can, but before you do let’s look at another thing optical timing gates cannot do:
They cannot measure instantaneous velocity of an object as it breaks the beam. This usually is not a big deal for the stop signal, or any of the splits, but it could sill lead to errors. For instance, the first thing to break the split beam might be a hand, whereas the next gate may be triggered by a knee, the stomach, or even a butterfly. However it is a HUGE deal for the start!
Because the beam is so small and the sensor so sensitive, postural sway or a billowing clothing is enough to trigger it. Therefore athletes cannot start right up against it.2
Typical test protocols, and manufacturer recommendations are to set the starting line at least 0.3m behind the timing gate. Some even go as far back as 2.0m, but the most common distances appears to be 0.5m. How far back does not really matter because regardless of the distance, you are unable to account for what happens before the first beam-sensor is triggered. And it could be a lot.
How much time did it take them to react and/or decide to move before they actually moved? Reaction time is critical to sports performance, so it should not be ignored. Distinguishing between reaction (cognitive) and movement (athletic) speed may be of interest, in which case a second test will be necessary. Either two perfectly targeted tests; or one that includes both measures, then a second one that singles out just one (which can be subtracted to get the other). Optical timing gates cannot measure either because they cannot measure when movement begins.
Once they started to move, was it immediately forward or was it proceeded with some sort of countermovement (CM)? Lab test show that even a subtle lowering of the centre of mass takes 0.2 seconds. A false step is another type of CM, where one foot (more often the lead foot, but can instead be the trailing one) is lifted off of the ground and replaced in approximately the same location takes even longer than a simple lowering of the centre of mass 0.3 seconds or longer. There is consensus that a CM is costly over at least the first few steps. Some have argued that the greater propulsive output following a CM results in better speed after 5 or 10 meters, but those researchers used optical timing gates placed one meter in front of the athlete to start the clock, rather than measuring their total time to move. This is nonsense and an example of bad science where they are not measuring what they intended to measure and drew erroneous conclusions. Sprinters and their coaches strive to explode concentrically out of a loaded stance, without any drop or wind-up, because that is the fastest way to run a race.
False steps, where the first step is backwards, are even worse —three times worse! Not only have they wasted time, they have also gone the wrong way, and have even farther to run to get to the finish line. They will obviously loose the race. With timing gates, though, it will appear as if they are faster than when not taking false steps. Consider the average set-up for a 10m test, where the start gate is placed 0.5m ahead of the start line. With the false step, elapsed time between meters 1-11 was measured. This should not be compared to elapsed time between 0.5-10.5m, because acceleration3 is so high over the first few steps of a sprint. Everyone should be faster from between metres 1 and 2, than they are between metres 0 and 1m, and by extension faster over 1-11m compared to 0.5-10.5m. A false step gives a flying start. It provides at least twice as much time, distance and steps to generate forward speed before triggering the start gate. Optical timing gates are oblivious to all of this,4 but it certainly matters in a game when racing an opponent to the ball.
Timing gates have been validated for research. Their measurements of time correlate strongly with time measured with other devices. Subjects tend to get the same results on repeated tests(without training interventions between sessions), and they are sensitive enough to measure fatigue or training improvements within subjects. They are valid within individuals but not between subject. That is because without deliberate interventions, most subjects will employ the same strategy to start each run, however optical timing gates and (often) the human eye cannot identify which strategy each subject uses. It is therefore invalid to make comparisons between two players using only optical timing gates to collect data.
| Player | Start Technique | Intended Distance (m) | Actual Distance (m) | Measured Time (s) | Actual Time to run 10.5m(s) | Finishing Position in Race |
|---|---|---|---|---|---|---|
|
A |
Drop |
0-10 |
0.5-10.5 |
1.8 |
2.1 |
2 |
|
B |
Concentric |
0-10 |
0.05-10.5 |
1.8 |
1.9 |
1 |
|
C |
False Step |
0-10 |
0.5-10.5 |
1.8 |
2.2 |
3 |
|
D |
Negative Step |
0-10 |
1.0-11.0 |
1.8 |
2.4 |
4 |
Despite this, a few researchers took the leap from intra-subject validity to assume inter-subject validity and the peer reviewers missed it. So, some others took that as a sign that it was okay to do the same. Then a couple of studies came out where they were unable to detect a difference in 10m sprint time between professional and amateur rugby (and soccer football) players when using optical timing gates, and the authors erroneously concluded it was because there was no speed difference. This led to coaches neglecting the need for speed, and stagnating (or even reversing) player development.5
In reality, speed is the thing that differentiates pros from amateurs6, winners from losers7, elite from sub-elite8, 1st league from 2nd league9, starters from reserves10, and all-stars from the average11. Sometimes it is also strength. Occasionally it is also body composition. Yet it is always speed.
This has been the case for decades, in hundreds of studies examining nearly every team, combat and racquet sport. These differences occasionally have gone undetected when researchers (or coaches) have used optical timing gates. This is because optical timing gates cannot measure what the researchers (or coaches) are trying to measure and cannot be used to compare between individuals.
There are more appropriate methods for measuring speed. As mentioned earlier in this article, optical timing gates are imperfect which is why they are not used at top level athletic competitions. That being said, they are still quite good for capturing split times and the finish line, so they can still be useful there. Accurately incorporating the start is the tricker part.
Using a pressure switch is an improvement, but is far from perfect. It is a simple upgrade, as most manufacturers already provide them as an option. The athlete starts with a hand or foot on a switch which, when released, starts the timer. For some sports, there is a lack of specificity in a starting stance that maintains pressure on the switch. They also may be prone to breaking from being stepped on or kicked. Most critically, they are unable to capture premature drops or movements of the other limbs. False and negative steps are only detected if the responsible limb is on the trigger. They also neglect reaction time.
Providing a visual or auditory start cue for the athlete that simultaneously starts the timer instead seems like a relatively straightforward solution. A light or horn that is triggered (by the researcher/coach) with the same switch that starts the timer might be adequate. More realistic cues (such as reacting to an opponent’s movement) are trickier to reproduce in a way that can be integrated into the timing software, but is not impossible. These approaches will measure a sprint time that includes cognitive (recognition and reaction times) and athletic speeds, though it will not distinguish between the two. If you care to know the contribution of each, then do another test that measures time from that same signal to movement initiation from the same stance.
How do you detect movement initiation? Force plates are great for capturing premature movements, however identifying the onsets of different movements is a bit subjective. Their tops don’t match any sports surfaces, and they are thick so have to be built into the floor. They are also very expensive for what you get. For these reasons, they are usually confined to research or for high-performance training centres where they can remain imbedded in the floor of a testing area.
High speed (HS), high definition (HD) video is probably the best method. The iPhone camera and some third party Apps have been validated for this purpose. It takes a bit of time to identify the start and end frames, but this also serves as a tool for giving feedback to the players about their stance and movements. Integrating the camera with timing gates at the finish line will require some custom software.
In fact, you can do away with the optical timing gates altogether! Record the whole sprint (or punch or swing) in HS HD video and identify the frames corresponding to the start, end and as many splits as you want in the same way as you did for the start. Just remember Pythagoras when marking out distances, and always calibrate the position of the tripod. The only real downside is that, for some bizarre reason, additional software is required to calculate movement time because Apple’s inherent camera software will not count the frames in a clip or assign a timestamp to them, even in the metadata.
Maybe this isn’t a problem, but it annoys me.
–CG
1 Technically it would give average velocity, because it considers the time for displacement from gate 1 to gate 2. The actual path traveled (which is unknown in this testing scenario and assumed/hoped to be a continuous straight line) is Distance, which when divided by time = speed. It’s arguable which is more relevant in sport and depends on the scenario and perspective. If a running back gets a hand-off at the 40 yard-line and runs all the way to the end zone, going a straight line perpendicular to the line of scrimmage would get her there in 4.6 seconds and give the highest velocity measure. More often, she will have to take a longer path to the end zone in order to get around opponents, moving at speed but taking more time than just sprinting straight ahead. This would give a low velocity, despite the similar speed. For two basketball players chasing down a loose ball, velocity matters because taking the most direct path will get them there faster.
2 It is also not visible light which makes it even harder to attempt getting close.
3 Which is defined as the rate of change of speed.
4 Apparently so are some coaches and researchers that erroneously employ these devices.
6 Wells, C. M. et al. (2012). Sport-specific fitness testing differentiates professional from amateur soccer players where VO2max and VO2 kinetics do not. Journal of sports medicine and physical fitness, 52(3), 245.
7 Ramos, Sérgio, et al. “Differences in maturity, morphological, and fitness attributes between the better-and lower-ranked male and female U-14 Portuguese elite regional basketball teams.” The Journal of Strength & Conditioning Research 34.3 (2020): 878-887.
8 Milić, M. et al. (2017). Anthropometric and physical characteristics allow differentiation of young female volleyball players according to playing position and level of expertise. Biology of sport, 34(1), 19.
9 Vigh-Larsen, J. F. et al. (2019). Fitness characteristics of elite and subelite male ice hockey players: A cross-sectional study. The Journal of Strength & Conditioning Research, 33(9), 2352-2360.
10Sell, K. M. et al. (2018). Comparison of physical fitness parameters for starters vs. nonstarters in an NCAA Division I men’s lacrosse team. The Journal of Strength & Conditioning Research, 32(11), 3160-3168.
11 Hedlund, David P. “Performance of future elite players at the National Football League scouting combine.” The Journal of Strength & Conditioning Research 32.11 (2018): 3112-3118.