Untold Arsenal on Twitter @UntoldArsenal
By Gordon Haverland, P.Eng.
Some of us Arsenal fans have noticed that Arsenal is using GPS
vests in the friendlies this summer (and at least once – at Barnet – last
year). We may have personal GPS, or GPS built into our cars. How does this help Arsenal? Can this help me understand what GPS is doing for me now?
I am an engineer who has always been involved with lots of mathematics and statistics. For about 6 years (1996 – 2002), I spent my life using GPS and geostatistics to help farmers in Alberta, Canada.
When I started working with GPS, a feature called Selective Availability was in effect, which scrambled the system such that positioning errors of 100m were relatively common.
At the same time, it was fairly common for GPS solutions to suffer “blunders”.
In GPS, a person is interested in the distance between the satellite and the receiver, and this distance is divided into two parts: the integer number of wavelengths of the radio signal, and the fraction that is left over (usually called phase).
A blunder happens when the phase is correctly calculated, but an error in
the integer number of wavelengths is made.
Precisions of 100m weren’t of much use back then in farming (and wouldn’t be useful in football), and so we actually employed something called Differential Global Positioning System (DGPS).
Over a precisely known location that was reasonably close to where we were working, we would set up a GPS antenna at some known distance above the ground. Typically we used formal survey markers. And we set up a little computer at this location which would be constantly comparing the location via GPS with the known location (to a couple of centimeters), and calculating an error vector. This error vector was transmitted over a radiomodem to
our “roving” GPS antenna.
By this means, we were able to obtain precisions on the order of 10cm (in general). But we still observed the occasional blunder.
With Selective Availability turned off now, most people should see positioning errors of 5-10m. Back in the late 90’s you could subscribe to satellite services, which would produce DGPS estimates on the order of 50-100cm. These kinds of corrections won’t get much better over time.
I would imagine that Arsenal is setting up their own GPS base station at the field, and calculating corrections. This would also be a good place to receive data from each player and store it on the computer.
Since the early/mid 90’s, the maths has improved, reducing the chance of blunders. Even back then, GPS was commonly solved via methods involving the Kalman transform. Do you remember FM radio receivers talking about phased lock loops? That’s an example of a Kalman transform.
In any event, in a Kalman transform you can tie together related information to stabilize the solution. Handheld GPS receivers have had atmospheric pressure sensors (nominally measuring changes in elevation), and magnetic direction indicators (compass) for quite a while, having gyroscopic information is also useful.
Our (farming) DGPS setup was measuring positions once per second. I am going to guess that the GPS setup for Arsenal is probably running 50 to 1000 times that fast. If two players collide, or a player collides with some other object (for example a goal post), these higher sampling rates would allow someone to get a fairly good idea of the collision forces. Having video evidence will be very useful in that circumstance.
The Arsenal.com article mentions that heart rate information is gathered. For me, I would want the gyroscope information, accelerometers and possibly body temperature. The article doesn’t mention whether the data is stored on the athlete, or broadcast over radio to a central receiver/computer.
Storage like flashdrives are quite small, but I would think that an important criteria is to minimize just how much extra mass is present in that vest, and whatever mass is there should not be a hazard to the player in the course of the game.
If the information is broadcast back to a receiving station, I would guess it is encrypted. I don’t think Arsenal would want anybody getting copies of the information. How much equipment is in that vest might be reduced if the GPS equipment on the player is sending back nominally raw information, which would have to be post-processed in order to convert it to positions (and orientations in space).
If a person just wants to know how far a player runs during the course of the game, trivial analysis of the GPS information will probably be pretty close. If you are trying to find out how much force any of the muscles working are exerting, things get more complicated. You need to model how the player’s body is moving, not just assume the GPS antenna is some constant distance from the ground in some known relationship to the athlete’s two feet.
Untold Media: Launch of new media section, and we need your help
Hi Gordon
Nice article.
I just had a query. I had read that the best use of GPS for Arsenal was for injury prediction.i.e. If the athlete used his front part for running then he was fine. But if he was running more flat footed he was more likely to get injured.
Can you please tell how. Thanks
Abhishek
There was a much more in-depth article about this system on the Arsenal website last summer and an interview with Gazidis. It said then that it would take a long time to collect enough data to use the system to prevent injuries. My question therefore is: when we buy/sell a player, do we then buy/sell the data too? Otherwise it’d take a long time to build up a profile of a new player. But also it’s a way of making it difficult for a buying club. Think of Nasri.
@Kristoffer: Not many clubs use the GPS system currently so I doubt that there is any transfer of data between clubs.
From what I know of GPS (I’ve worked with them for automotive insurance), when sufficient data is collected, it is possible to create models for player movement and analyse how much stress is being placed on each muscle, what are the fatigue levels and what is the risk of injury. There was an article on Arsenal.com explaining this to some effect last year. Currently, these vests will only be used during friendlies and training sessions since they are not permitted under EPL rules.
Thanks Gordon for this information.
And I do have a few questions if you could give us more information.
Could this be used a solution for let us say deciding an offside position? Could all the players (of both teams) on the field have such a vest and then the computer could decide if a player is off or on side?
Could a ball be given a GPS chip to see if it crosses the line completely or not?
Just a few questions that I have after reading this
The only problem would be, if the info on the GPS is send via radios or computers, is it possible that opposition teams can intercept these frequencies to use the info for their own use, to identify and target our tired players??
I am asking this bcoz, radio frequencies and computer networks can be intercepted and be hacked into by any computer nerd….
Maybe News of the World hacked them… 🙂
Anyone remember the times when football was a game and science one of the pesky things that interrupted a good old kick about?
Sports results 20?? :
Arsenal Androids 4 v Man U Robots 0
Entertainment value – zero
hi Walter. What are the chances of other clubs hacking into this signal to get information about our players and their performance levels?
Walter,
we already have the technology to decide whether players are on side or off side purely using an array of video cameras.
What we are so blatantly lacking is any political will to implement technology, since it becomes almost impossible to fix matches with technology deciding whether goals count or not.
Cheating will always be a part of sport but in sports where technology is used to reduce it to a minimum the incidence of official match fixing has dramatically dropped. Association football could clean up it’s act in a massive way by copying the better parts of Rugby football’s rules and regulations.
@woolwich p
sepp “bribes, what bribes” blatter is pushing 80. the sweet f.a. board can’t be far behind. these guys don’t even realize the cell phone has been invented let alone care about technology, so we are in for a long wait.
GPS systems are tracking a single point on the body, that being where the GPS antenna is. One usually assumes that how the GPS antenna is “mounted” doesn’t change with time. What we get from this is position in 3D over time. If we look at the differences in position over time, that gives us velocity. If we then look at differences in velocity over time, we get accelerations. If we know accelerations and the mass of the body (here an athlete), we can calculate forces. Which is part of what we want. If we want to figure out how much each muscle is contributing, we either need sensors (accelerometer most likely) at convenient places, or we can digitize video frames to come up with the position of each joint over time. From what I’ve read, the next derivative with respect to time from acceleration (called jerk) is important in bone injury and healing. In the absence of collisions, we might only be interested in jerk in elevation. (We being people in general.)
Offside or goal line? GPS would give you an answer in obvious cases, but as the case was obvious, GPS wouldn’t have been needed. The problem with either, is that “shape” of both the athlete and the ball change (the athlete more so). Is the point on the athlete involved in offside, the point on their body closest to the goal line (which could be a hand), is it the part of their foot closest to the goal line, is it the part of their foot in contact with the ground that is closest to the goal line? The same sort of problem happens on the goal line. The sensor is most likely suspended at the center of the ball, when the ball is spherical. Ignoring that when the ball changes directions or is hit, the sensor will oscillate around this center for a while (damped vibrations), all the sensor really says is that the center of the ball got to some position. If a player is on top of the ball, they are probably squeezing the ball out of a spherical shape, making it bigger in the horizontal plane. And all of us are largely (salty) water (which conducts electricity), which absorbs the radio waves needed for GPS quite well. GPS might help when the ball is visible, but things like video work for that as well. I don’t think it is practical for offside or goals though. Perhaps someone has a way?
I would think the radiotransmissions to be saved would be encrypted, much as a VPN. Anybody could record the tranmissions, but they would need to decrypt them before they would be of much use.
Science has already helped most sports by allowing better training methods and faster recovery from injury.
Sorry for being “late” in responding. I’m in western Canada, and I haven’t even had morning coffee yet.
@mike in Atlanta
I’m sure they’re well aware of the technology that could be used to improve refereeing in football. And it scares the crap out of them.
I don’t think there are many places where technology would help, or at least not for a reasonable price. Offsides? I think video would work, but you need cameras that are aimed perpendicular to the sidelines, always tracking the second last person. If you have some number of elevated cameras viewing the entire field, you need to be constantly calculating where the ball and players are. And situations where no camera has a good view of what is happening will always happen. What kinds of games do you apply this too? Just the top mens professional leagues?
@Gord,
Two things:
Firstly, the camera mesh you speak of exists already for any match which is monitored by ProZone although I believe the analysis is done by humans rather than machine vision in that case.
Secondly, what the GPS system is looking for is deviations from normal biomechanics as determined by acceleration. Players that aren’t running ‘healthily’ will decelerate and accelerate more at footfall than a healthy player. If the problem is localised in one leg or is caused by another ‘sided’ injury (even if it isn’t noticeable) will produce unusual acceleration vectors as part of their running cycle. You don’t actually need to do much more than take the second derivative of position, merely compare that with the established norms. People always assume that these things are more complicated than they need to be.
I think there are places where science and technology can help, but I don’t think that adding technology to the game is something that will ever be 100% effective. Lots of people talk about video for goal-line incidents. Somebody probably installs a single camera per goal area, and it may help in 80% (just a fabricated number) of the situations where a referee and lines-person didn’t work before. But in those 20% where it didn’t work, there will be some situations where it doesn’t matter how much someone analyses things, there is not enough data to make a determination. A referee could end up halting play for 20 minutes trying to get some magic out of the video, and it may never come.
Where I think video can do the most good, is detecting off the ball incidents. But that is a disciplinary problem dealt with after the game is finished.
I have a long standing interest in athletic first aid, and for a while I taught weightlifting (volunteer). And I do read stuff like how NASA approaches fitness problems. I don’t think what you think are established norms, are all that established. I think that with time we will continue to find reasons for observed variability that are explained by genetics, environment and past history. But, it is a place to start.
Gord,
Those established norms are established by the players themselves. The whole system is relative. I’ve been working on establishing a deviation-from-personal-norm system for medical purposes for awhile now, after looking at some data collected from very abnormal subjects. The approach of looking at what is “normal” is fundamentally flawed when it comes to humans (and racehorses…)
Woolwich I think we agree on that. I’ve been playing statistics a long time, and I wish the word “normal” never would have become associated with Gaussians. 🙂
Mike in Atlanta,
Woolwich Peripatetic is spot on re. the video technology scare. Cleanie Platini and Septic Bladder, just a week or two ago both were reported to have come out dead against it, saying it was unwanted in football and that more than one ref watching the goal line would do. This was said, mind you, the day after – repeat the day after – the Greek corruption crisis broke out and 41 “suspicious matches” were flagged leading to what’s now up to 84 arrests, including referees and club owners and police as well. And here’s the kicker: it was UEFA reports, the day before, that led to the arrests. The arrests, in UEFA/FIFA speak are “evidence” that they are on the case, and the system “works.” Rather than video technology is needed to ferret out the corruption that had resulted in 41 flagged matches. This is the true fear of technology – the fear of a more level pitch and hands caught in the cookie jar – that is afoot.
Interesting article, not so many blogs write about the use of technology in sports. My thought about this is that it can be very useful for managers, coaches, players to know about their performances and capacity, thus injury may be earlier detected. However, the main braking factors are the policy makers as well as the technology itself. These two factors can be a chicken-egg problem. As long as the technology cannot demonstrate reliability, then the policy makers will be reluctant to employ it. And without the support from policy makers, there will be less money involved, thus less effort will be taken to better the technology. If, for example, FIFA really wants to apply technology, they can create a competition, something like Lunar X Prize, open to everyone to explore and develop a technology that can be really applied in Football.
Good thing is, the technology can already be used in training as there is no restriction or regulation from the policy makers. Thus, it’s up to the clubs whether they want/can use it or not. Other sports are already using several technology, e.g. GPS, sensors, vision-based player tracking, to analyze the on-field behavior of the player during training. In the university where I am studying, there is a cooperation between mathematician, sports, and engineering to apply video tracking and sensors for analyzing games of handball and basketball. Perhaps in some years ahead, we will see technology in football. One technology that I think can be immediately applied is video technology as we have seen it many times in football and it is used as a tool for helping a referee, not for replacing them. So bring video technology to football.
AnT.
Crazy as it sounds (and it is!), in the EPL, the referees do not want the help of video technology, or they wouldn’t be there. Welcome to Untold Arsenal…
Woolwich, perhaps you aren’t following this thread any more? Want to discuss things offline? You seem to have complementary knowledge to mine.
I still don’t think video playback is useful for assisting referees. I am sure most people can go searching for situations in televised games where video replay would have provided better information to the referee. I have no doubt they exist.
If football was to institute video playback, whose cameras are they going to use? They will NOT be using commercial broadcast cameras, as they don’t have _control_ of the data.
Where do you position the camera, or multiple cameras? Lots of other details about observation.
I still think the best use of video is in disciplining off-the-ball incidents.
But to have a referee go to a booth to review video during a game to decide something which was controversial, means that the review time may never end. No matter how many times it is reviewed, no positive determination can be made. And long delays in play is something that FIFA has been against.
Sure, bring in cameras. If the referee decides for a review, go for it. If after 30 seconds, it isn’t obvious what the decision should be, go with the original decision by the referee. But football isn’t like gridiron in North America, where things are stop/start (mostly stop).
How does one demonstrate reliability? I think that in terms of league play, GPS is never going to be involved. In order to have GPS equipment on an athlete, means that there is an enhanced possibility of people being injured, because they are components on the athlete which could result in injury to the athlete or a competitor. I do think friendlies are a way for teams to find information beyond what they normally can
If there was a serious collision in a game where things like GPS vests were allowed (such as a friendly), the GPS will provide information useful in studying the serious collision. And this includes having sensors not involved with GPS (gyrometers, accelerometers, …). But, the idea of having friendlies is somewhat serious play. Anything that happens is closer to “real play”. A person still needs video in order to model the body during the collision. And there is always the possibility that the extra instrumentation is a contributing factor in injuries.
In terms of monitoring the core temperature of an athlete, I think swallowing a probe before game time makes the most sense. If the athlete is wearing a GPS vest meant to send data to an external computer, it is probably pretty easy to juggle transmission strength so that the vest picks up the internal temperature.
Are we going to be able to miniaturize GPS to the point where a person could swallow a capsule (instead of wearing a vest)? Probably not in the near term.
The purpose of friendlies (IMHO) is to experiment. But the experiments have to be controlled. Whether it is personnel, tactics or things like GPS. One hopes that any experiment in a friendly, produces useful information.
Gord
EPL’s Scudamore is rapturous at adopting “goal line technology” for 2012-2013, as a fantastic step forward that snuffs the idea that something about the EPL is “broken.” So, a crumb from the table to handle a tiny fraction of the disputes, to pretend that something significant and so serious is being done about the state of refereeing, etc. The only serious aspects of all this is (a) who gets the profitable contract for this technology and (b) that it’s a serious and intended diversion from the growing outcry to improve referee performance that has been well-documented last season on this website. Today’s Guardian covers the EPL boss’s faux solution to the problem: http://www.guardian.co.uk/football/2011/jul/19/premier-league-goalline-technology
Gord,
A ref-in-the-booth could have a specific remit of 1-3 types of in-game reviews that could overturn calls made just below on the pitch. Here’s one: Review of seeming contact (to nullify dives) that results in penalty kicks. This is quick, not long: these multiple camera angles already appear instantly on the broadcast TV for viewers; and many are quite clear as to whether there’s really been contact or not. If not, then too close to overrule.
Now you are oh-so-concerned about “long delays” I submit that there’s a LOT longer delays caused by the diving and the obligatory drama of on the ground writhing and the earnest entry of the stretcher bearers and lo, the sudden phoenix-like sideline resurrections of the “injured” players. And, my point: that a ref-in-the-booth who reviews such behavior when it’s the basis for penalty kicks would (a) prevent the worst injustices (such as when Riley/ManUre/Fergie did to us) that mar football; and (b) SAVE time by discouraging the number of crap diving scenarios that chronically slow up the games – especially when a faux dive, caught by the ref-in-the-booth, would result in a card – red, yellow or purple.
I am very skeptical of invocations of wasting time when arguments like yours invoke the worst-case scenario of endless reviews, whilst ignoring situations like the one I raise above (as just one of several possible scenarios) in which video-review would actually SAVE (or NOT WASTE) time, your main concern, it seems, when compared to getting the call right. To me it is far worse to have a game turn on a mistake than to interrupt the flow that is already being interrupted, and not infrequently Gord, by behaviors like game-changing dives.
Besides.. who says the clock CAN’T be stopped? Yes we don’t want a stop start game..But look at Rugby. It is a nonstop game and free flowing. But they stop the clock to determine through a video ref whether the ball crossed the line on a try or not. It might take a minute or two, but the attempt is to get the call right. If it is inconclusive, the decision in field stands.. (Also, time keeping should be separate anyway, since at the moment the ref has too much power to determine how much time he should add)
Another simple enough thing to do is put a mic on the referee like rugby does, so that everyone at home, and in the crowd can hear what the referee is telling the players. That sort of scrutiny on the ref is more than compensated by the fact that none of the players are allowed to show ANY dissent. All in all, the object of a sport has to be to promote fairness. That is most certainly not the case in football right now.
Shard, Gord, Laundyender,
yes!, and there actually is more often than not suspense created by waiting for the call to set it right or not. It is not at all necessarily the dead time that the seeming purists invoke. Drama is drama and football is often high drama, and we do love it dramatic. Waiting on a call that can get it right is not in itself at odds with an exciting match, and can add to the fan’s experience. It’s just bogus to paint this delay as an intrinsic bore, or some sin against the sacred game that’s already being so tarnished nowadays.
BTW, please check out Laundyender’s link (posted today, but on the ) to the Telegraph’s major article on match-fixing: telegraph.co.uk/sport/football/8648454/Global-match-fixing-investigation-claims-major-scalp-as-Wilson-Raj-Perumal-is-jailed-for-two-years.html
Bob, I agree that goal line technology is less useful compared to other technology that can improve referees’ decision making. But we can just hope that it’s a small step towards more application of technology in football.
Gord, I think it is the reason that such vests are not allowed in real games as there is a possibility that they endanger players’ safety. A possible idea is to have the sensors as small as possible and self-sustain so that they can be placed on shirts. Well, for sure I would like to have such a replica if it is available. 🙂
I’ve been remiss in following up.
@bob
Part of the problem with reviewing things, whether it is at game time, or afterwards, is the frame speed. I gather the strongest kick recorded in soccer is 139 mph (62+ meters per second). The ball (when spherical) has a diameter of 0.22m. Hence, we need frame rates (not interleaved) of 282 frames per second to catch goal events when no barriers to viewing an incident are present. Normal video I think operates at 24 fps. And really, we probably want at least twice this fast.
Nullifying dives is probably in the same situation. I have seen slow motion on TV of things, and there is no way to tell. The problem is the frame speed. We probably don’t need 300 fps to detect dives in video, but if we need 300 fps to detect goals (maybe), then any of our monitoring systems need to run at 300 fps.
Hence, we cannot take a feed off network TV (if available) to provide good video evidence. We need to have special video running at high frame rates.
Almost half of my time involved in football has involved athletic first aid. If someone is hurt, I want as much time as it takes. I have been at fields where someone has suffered double leg breaks. And the league I was involved in had one person die from a cerebral hemorrhage (indoor football, head contacting rubber coated concrete). In the case of the double leg break, having paramedics on site would have helped. For the guy who died from the cerebral hemorrhage, it wouldn’t have mattered if paramedics were present (and the athlete in question I think was a dentist in regular life).
If there is a legitimate injury, I want all the time it takes to treat it. But, if play is delayed, all the players will become colder. And becoming colder means becoming more susceptible to injury. Fine, we have a 5 minute delay to study some dive (which can be dealt with in a post-game disciplinary review). Some athlete becomes too cold, and tears a hamstring muscle on restart. If you are offering 30 second reviews at maximum, I will take them. If a video review takes more than a couple of minutes, I don’t want it. It is not worth the possibility of a cold athlete injuring themselves seriously because of the pause.
@AnT
I think the best we are going to do in the near term, are sensors a person can swallow. Which puts them a reasonable distance from where any collisions can happen. And some time after the game, they come out.
@bob
Lots of people think drama. I’ve spent entirely too much time patching athletes (including myself). Leave the length reviews for baseball (which really isn’t a sport, with energy expenditures not much different from chess).