the hawk-eye con

Discussion in 'General Pro Player Discussion' started by nousername, Dec 9, 2011.

  1. nousername

    nousername Rookie

    Joined:
    Sep 26, 2007
    Messages:
    317
    Does it bother anyone else that the Hawk-eye system uses 12 HIGH-SPEED CAMERAs, yet challenges are based only on the digitized 3D ball-trajectory estimation?

    Why not use the figiggin' high-speed video??? at TWELVE angles!!!

    Hawk-eye has brilliantly conned the ATP and Grand Slam events into thinking their system is needed for challenges.

    Is it not obvious to everyone that an umpire with access to 12 high-speed camera angles could do a MUCH better job than a computer generated estimate?

    Shouldn't the high-speed video with a human at the wheel be used instead?

    Especially in "unique" situations like this:

    http://www.youtube.com/watch?v=IhUmGiT5GqY&feature=related

    The ball clearly hit Petzchner's leg, and the point should have gone to the Poles. Playing back the 12 high-speed video feeds surely could have cleared this up.

    The digital 3D ball-trajectories are good for commentary and teaching. But the video feeds that generate those trajectories should be the source used for resolving challenges, NOT the digital 3D trajectory.

    this has been my pet-peeve with the pro game ever since the advent of hawk-eye ...
     
    #1
  2. Sentinel

    Sentinel Bionic Poster

    Joined:
    Jan 21, 2008
    Messages:
    31,247
    Location:
    Somewhere under the weather ;)
    I believe they've conned cricket as well.
    Does Hawkeye use the wind at each and every point in its estimation ? Because I know that FedEye does. That's why Fedeye often conflicts with Hawkeye.
     
    #2
  3. Xizel

    Xizel Professional

    Joined:
    Feb 13, 2011
    Messages:
    1,404
    I remember when Tsonga saved a ball before the second bounce and the umpire thought it was two bounces already. Only the commentators' replay saw it. Although Federer ultimate won the point, the referees could use replay feedback.
     
    #3
  4. Talker

    Talker Hall of Fame

    Joined:
    Oct 4, 2007
    Messages:
    2,990
    The ball did directly hit his leg, he was asked but he denied it. :shock:

    It would take too much time and the crowd wouldn't be involved so I doubt it'll be used.
     
    #4
  5. MajinX

    MajinX Professional

    Joined:
    Apr 12, 2008
    Messages:
    1,185
    Location:
    Toronto
    Im dont think the video you posted was a good example... coz he wasnt challenging whether the ball was out or not, he was trying to tell the umpire the ball hit his foot which the umpire did not see happen.

    And from my understanding the 12 highspeed cameras are what is used to generate the 3d "estimation" visual we see when players challenge. I could be wrong about that tho.
     
    #5
  6. nousername

    nousername Rookie

    Joined:
    Sep 26, 2007
    Messages:
    317
    also, in those close calls where Hawk eye calls it out or in by a matter of millimeters. in those situations, it would be much better for everyone fans, players, umpires, etc. to see the 12 video fees in high-speed, and then let the ump make a judgment call. everyone would be happy with that. tennis is kind of based on the honor system, ALL calls should ultimately be based on a human's judgment rather than some overly technical digital system.

    the system as used is a joke.
     
    #6
  7. nousername

    nousername Rookie

    Joined:
    Sep 26, 2007
    Messages:
    317
    i'm not talking about the "challenge" made in the video, that was just a concession by the ump. he told matkowski the only challengable call in the point was the "winner" by Petzchner. so they challenged it out of desperation. i'm not talking about that challenge.

    my point was that the 12 high-speed videos should be available for ANY kind of challenge. Matkowski should have been able to challenge that it hit Petzchner's leg. the 3D hawk-eye system can only make COURT calls, where as a human watching a video playback certainly could see it went off his leg.

    yes, you are right. the 12 video feeds are just to track the ball and generate a 3D trajectory. then the trajectory, and trajectory alone, is used to make the hawk-eye challenge calls.
     
    #7
  8. OddJack

    OddJack Legend

    Joined:
    Jan 1, 2009
    Messages:
    9,663
    Location:
    South
    Hawk eye is pretty accurate, maybe not 100% but to millimeters.

    Video cameras will not show the ball mark on the court, simply because the ball itself is always wider than the mark it leaves on the court, which is the point of impact, which is what counts.

    Plus, They want to leave umpires and any other humans out of it for arguments sake.

    The cameras and the computer are pretty accurate. There is simply not enough time, and sense, for umpires to sit down and watch 12 different angels and then decide if the ball was in or out and then leave the door open for arguments, was he right? Or wrong? As someone else pointed out Layhani, one of the bests, got the number of bouncing wrong, and you want him to have millimeter precision.
     
    #8
  9. nousername

    nousername Rookie

    Joined:
    Sep 26, 2007
    Messages:
    317
    it would be just as fast as hawk-eye is now. you can even show the nice fancy 3D video for the audience and commentators, but let there be one judge in a booth that watches the video feeds. let him instantly plays back points to make the "official" call when a player challenges.
     
    #9
  10. woodrow1029

    woodrow1029 Guest

    If it were actual video, there would still be the human element of subjectiveness as to actually at what exact point contact was made with the ball and the ground.
     
    #10
  11. nousername

    nousername Rookie

    Joined:
    Sep 26, 2007
    Messages:
    317
    it IS accurate. that's one of my other issues with it. it is too accurate for the game, it defeats the essence of the game.

    tennis is generally an "gentleman's" game, it's always been based on honor. at all levels (except the pros) you make your own calls. the challenges would sit better with everyone, esp players and fans, if the actual video was used. in general the rule of tennis is "if you can't tell it's out, then it's in". a system that measures down to millimeters too good; it is beyond the scope and nature of the game. if it's not visibly out, then it's in. everyone is happy with that, and that is why you MUST have a human-in-the-loop on tennis replays.
     
    #11
  12. OrangePower

    OrangePower Hall of Fame

    Joined:
    Sep 7, 2007
    Messages:
    4,073
    Location:
    NorCal Bay Area
    If you've ever watched the replay system in action in football (american), then you know how excruciatingly long it takes for humans to review camera footage from various angles in the event of a challenge. Tennis has become slow enough as it is. Hawk-eye isn't too bad in the sense that you get a decision pretty quickly. Having an official watch the video replays from various angles would be slow death.
     
    #12
  13. nousername

    nousername Rookie

    Joined:
    Sep 26, 2007
    Messages:
    317
    yes, and no.

    actually, hawk-eye makes all kind of assumptions about the ball such as it's elasticity, the thickness of the felt, how much it compresses on impact, etc. it also has to map out the lines of the court, which it could be wrong about. hawk-eye then does some cool estimation stuff to put the ball on their court. that the point it's accurate to millimeters in hawk-eye's digital version of the court. it's not like the 3D trajectory get's over laid on the real court, the court itself is an estimate.

    nothing is more accurate than looking at a dang picture/video of the ball landing on the court. in a picture both the ball and the court are REAL.

    there are so many source of potential error, that a human watching high-speed video can make a much better call (not necessarily a "more precise" call in terms of distance in or out), but a "better" call that is inline with the essence of tennis.

    tennis is a game of honor, and i think the calls should be in the hands of a human.
     
    #13
  14. OddJack

    OddJack Legend

    Joined:
    Jan 1, 2009
    Messages:
    9,663
    Location:
    South
    Yes that's gentleman's game, but unfortunately Ms. Serena Williams plays it too, which is who started it all.

    There are too much of controversies, conspiracy theories, racial claims not to mention Money involved to leave it to gentlemanly procedures alone to decide the fate of the match.

    Expect to see it in soccer too. Blatter finally gave in to demands for Goal line technology for world cup 2014.
     
    #14
  15. Bobby Jr

    Bobby Jr Legend

    Joined:
    Jul 5, 2010
    Messages:
    7,499
    The Hawkeye system is pretty good for what it is intended for. It resolved line-call disputes quickly and fairly (as far as we can tell) but the other really good aspect of it is statistics calculating. The stats they derive from it such as average return point position, serve placement, height over net etc etc are fantastic additions to the entertainment package.

    I do agree in some ways though, and have said here in the past, that they should call the Hawkeye company's bluff and say they're going to also use high speed cameras at a future event to see whether the claimed accuracy levels are as claimed. I've seen, in the early days of Hawkeye, footage of a line-call which suggested the Hawkeye estimation was as much as 15cm wrong. (it was a straight on call so almost no side-ways blur with most conventional footage).
     
    Last edited: Dec 9, 2011
    #15
  16. nousername

    nousername Rookie

    Joined:
    Sep 26, 2007
    Messages:
    317
    don't you remember the mac cam? it's was quick and instant. it's be the same way.

    tennis is not like the NFL. in tennis the only issue is where the ball hit, period. the ball is relatively unobstructed the whole time, just 2 people and a net. in the NFL it is 22 guys plus 7 refs AND more importantly the NFL is plagued by rules. most replay time is discussion of the rules (i'm guessing).
     
    #16
  17. TTMR

    TTMR Hall of Fame

    Joined:
    Jul 24, 2009
    Messages:
    1,943
    As if player and umpire wouldn't argue over the same shot in a video still. I'm no physicist, but I suspect hawkeye has all kinds of technical advantages over video that we are not even considering.
     
    #17
  18. jokinla

    jokinla Hall of Fame

    Joined:
    Apr 24, 2011
    Messages:
    2,705
    Not to mention, this would be used as gamesmanship for sure. You can imagine, a big point, a player would challenge, knowing it would take several minutes to sort it out, in the mean time, killing a bit of momentum, or perhaps giving a player, several much needed minutes to rest, and anything that took that long could never be used between first and second serves, it would never work.
     
    #18
  19. slickerthansleek

    slickerthansleek Rookie

    Joined:
    Oct 19, 2011
    Messages:
    125
    Location:
    Sydney, Australia
    Have you ever watched a rugby league game? The way that we look at replays time and time again from all these different angles to figure out whether a try has been scored or not is a pain, and a real momentum killer. For every one or two points hawk-eye screws up, we get a fair result on a thousand of them. So I'm not too bothered, I'd rather hawk-eye get one wrong every now and again than get all of them exactly right but have a system in place which kills the flow of the game.
     
    #19
  20. Bobby Jr

    Bobby Jr Legend

    Joined:
    Jul 5, 2010
    Messages:
    7,499
    Yeah, the momentum killer aspect in league is all-but irrelevant if someone has scored as they'd spend 2 mins doing the conversion attempt anyway. If not, they still have to re-set for a goal drop-out or restart which is quicker, but usually a fair bit of time still gets eaten up regardless.

    The thing which is interesting about league video refs is they still manage to get piles of video ref calls wrong - often on benefit of doubt calls or obstruction in the preceding phase to a supposed scoring. Plenty of refereeing bias still exists in that system too - you only have to view any number of calls against the Warriors to see that.

    The area where league has really improved it's game management is the two ref system in the past 2 years (or is it 3?) and I agree totally... the positives of Hawkeye in tennis probably outweigh any failings in the system by more than 100:1.
     
    Last edited: Dec 10, 2011
    #20
  21. Crisstti

    Crisstti Legend

    Joined:
    Jul 8, 2011
    Messages:
    7,269
    Location:
    Chile
    I see your point, hadn't thought about that. It does have many advantages too though. It's quick (not killing the flow of the game, which is the main argument for not having replays in football) , it's entertaining to the audience, everyone can see the replay.

    Still think that the umpire checking the actual ball print on clay is better, so I don't see the need for hawk eye there...

    Didn't know that about the estimation of the court itself :|

    I think the goal line issue is one of the least problems, really. Some kind of replay for certain situations - penalties, red card offenses, goals disallowed for off-side - are way more necessary. After all one gets a goal line controversy once every 5 years or more, while the other situations happen all the time.

    Good point.
     
    #21
  22. mellowyellow

    mellowyellow Hall of Fame

    Joined:
    Jun 5, 2010
    Messages:
    2,584
    I think what bothers me the most, is that the system surely knows when balls are in or out in real time, why do the challenges even need to be made? I am not saying to totally rely on the system, but you have to wonder why should someone have to risk stoping a point to challenge when the eye already knows in or out? This could be visually done with a green and red light on the net post out of main visibility, and on the refs little computer.
     
    #22
  23. Manus Domini

    Manus Domini Hall of Fame

    Joined:
    Jan 10, 2010
    Messages:
    4,859
    Location:
    Jersey
    #23
  24. Zildite

    Zildite Hall of Fame

    Joined:
    Jan 27, 2011
    Messages:
    1,903
    So line judges would become obsolete? Unless you really did hide it well.
    Does it even work that quickly, I don't know.
     
    Last edited: Dec 11, 2011
    #24
  25. TTMR

    TTMR Hall of Fame

    Joined:
    Jul 24, 2009
    Messages:
    1,943
    Linespeople are probably unionized, so it won't happen. It is like having an electronic strike zone in MLB; it would completely eliminate the often random and inconsistent strike-calling of umpires, but the umpire union is drastically opposed because a large percentage of officials would be rendered redundant.
     
    #25
  26. pudelko

    pudelko Rookie

    Joined:
    Sep 9, 2008
    Messages:
    296
    Location:
    London Ontario, Canada
    Ive often thought about this myself. Even if its not real time, its definitely fast enough for the players to just wait a second or two (towel off?) between points to see the result of a questionable shot.

    Often times on TV when a player decides not to challenge the TV station shows what hawkeye would have shown anyway, so its seems safe to assume that the feature is available for all shots and the whole challenge thing with the dramatic sound effects and the umpire asking for a challenge is just there for the entertainment of the crowd. Not to mention all the extra time this takes.

    Tournaments that dont have hawkeye should fire all linespeople they have now and use that money to install the hawkeye system. :twisted:
     
    #26
  27. OddJack

    OddJack Legend

    Joined:
    Jan 1, 2009
    Messages:
    9,663
    Location:
    South
    The goal line problem is less frequent, and that's why they are starting off with that. There has been some stubborn resistance to use technology in soccer so this, if it happens, is a significant change.
    In one of the last world cup matches the ball had clearly passed the line completely and yet it was overlooked.
     
    #27
  28. Bartelby

    Bartelby G.O.A.T.

    Joined:
    Nov 14, 2005
    Messages:
    12,936
    Linespeople are usually volunteers or in casual employment, at best, so hardly unionized.


     
    #28
  29. Crisstti

    Crisstti Legend

    Joined:
    Jul 8, 2011
    Messages:
    7,269
    Location:
    Chile
    Well, that was just karma ;)
     
    #29
  30. nousername

    nousername Rookie

    Joined:
    Sep 26, 2007
    Messages:
    317
    two key points that you, and many others, are missing:

    1) these are HIGH SPEED video cameras, i.e. probably 1000 frames per second. There will almost never be a disagreement with that kind of time resolution. it's not a matter of one low-res image that may or may not have been snapped at the right time.

    2) there are 12 camera angles available to choose from. and tennis is NOT football or rugby, there's not clutter or obscurity of events. you have 12 angles to see ONE ball on a large open high contrast area. AND you won't need all 12 one time. with experience the umps will know exactly which of the 12 to choose for any one review. you would only need to look at 1 or 2 angles for a given shot.

    if you guys will take time to dwell on these facts you will come around. hawk-eye is an unnecessary element of the game. hawk-eye is a business and want to sell an awesome product. i love hawk-eye as a technology, but they've conned tennis. tennis needs their camera setup but not their 3D processing.
     
    #30
  31. Lsmkenpo

    Lsmkenpo Hall of Fame

    Joined:
    Aug 21, 2006
    Messages:
    3,915
    The one flaw with hawkeye is it uses a 3d model of the court, that is fine on a hardcourt which doesn't change but on a natural surface it is not as accurate. The court level is not completely flat on grass or clay and doesn't remain constant.
     
    #31
  32. woodrow1029

    woodrow1029 Guest

    I don't see how a system that has been proven to be accurate to around 1 millimeter, helps the officials get the call right, removes subjectivity on a very, very, very close call, and keeps the matches moving without argument from the players is a con.
     
    #32
  33. j00dypoo

    j00dypoo Rookie

    Joined:
    Dec 22, 2010
    Messages:
    275
    Location:
    Atlanta
    nousername is failing to take his own advice and "dwell on the facts." All the great points have already been made against you.

    1) can't stop tennis matches and review 12 camera angles and take 2 minutes to make a line call. Hawkeye resolves the matter quickly and efficiently.

    2) gives an ultimate answer. No human subjectivity involved which is great. You can't argue with a computer. It's like hitting against a wall - you will always lose.

    3) It is accurate. It's been proven accurate and it's only going to improve.
     
    #33
  34. Lsmkenpo

    Lsmkenpo Hall of Fame

    Joined:
    Aug 21, 2006
    Messages:
    3,915
    "According to the ITF criteria, any electronic line-calling system must be able to judge a ball in or out within 5 millimeters (0.20 inches). Incorrect calls are allowed, so long as they are not more than 10 millimeters (0.40 inches) off." :shock:

    http://sports.espn.go.com/sports/tennis/wimbledon08/news/story?id=3452293 Two British scientists call into question Hawk-Eye's accuracy
     
    Last edited: Dec 13, 2011
    #34
  35. LuckyR

    LuckyR Legend

    Joined:
    Jun 2, 2006
    Messages:
    6,404
    Location:
    The Great NW
    You clearly don't get it. The whole point of Hawkeye is to get an objective measure. Some things can't be known absolutely, most people can wrap their minds around that reality. What is truly unreliable is a person making a "judgement call", since people (unlike a computer averaging 12 images), have their biases.

    As to having an immediate answer instead of having everyone in the crowd look up at a screen and see the image in slo mo, you obviously aren't in charge of selling tickets, since it is a large crowd pleaser.
     
    #35
  36. Lsmkenpo

    Lsmkenpo Hall of Fame

    Joined:
    Aug 21, 2006
    Messages:
    3,915
    Point number three is highly debatable, hawkeye's average margin of error is 3.6mm, some may say that is not exactly accurate, hawkeye is not a real measurement , it is a model of what may have happened.
     
    #36
  37. pennc94

    pennc94 Semi-Pro

    Joined:
    Jul 26, 2006
    Messages:
    759
    Location:
    Maryland
    ^^^^
    I have been saying this for years.

    Drop the need for players to decide whether or not to challenge. They should be focused on the tennis (not deciding if a linesman makes an error).

    In addition to the red/green light, I suggest connecting a speaker that proclaims "OUT". I vote to have it use the voice of James Earl Jones.
     
    #37
  38. Lsmkenpo

    Lsmkenpo Hall of Fame

    Joined:
    Aug 21, 2006
    Messages:
    3,915
    ^^^^^^^^^^

    Hawkeye doesn't work that way, it is a statistical model of what may have happened on court calculating the balls trajectory, not a real time measurement of the actual shot. It is an accurate guesstimate.
     
    #38
  39. pennc94

    pennc94 Semi-Pro

    Joined:
    Jul 26, 2006
    Messages:
    759
    Location:
    Maryland
    Agreed it has its limitations (not being 100% accurate), but it appears it might get things more right than wrong as compared to the human eye. Has that comparison ever been made?

    Let's say that its margin of error is 4 mm. Then call LET when the mark (as estimated by Hawkeye) is within the margin of error.
     
    #39
  40. Magnetite

    Magnetite Professional

    Joined:
    Jul 7, 2011
    Messages:
    907
    One of the worst ideas ever.

    I don't see Hawkeye making terrible calls, but I saw lines people make terrible calls all the time.

    Umpires would also make terrible calls, and players would blow up. The player that got ripped off would then be too angry to play and tank the match.

    Terrible idea.
     
    #40
  41. nousername

    nousername Rookie

    Joined:
    Sep 26, 2007
    Messages:
    317
    i respectfully disagree, and it seems you are confused about the definition of being "objective".

    an ump making the call is an objective decision just like hawk-eye. they both base their decision on an interpretation of facts. the ump uses a visual image in his brain, whereas hawk-eye uses a visual image in a computer. the only difference is the amount of uncertainty in each process. both have uncertainty, but at different levels. (subjective is like "is wheat bread tastier than white bread?" that's a personal decision. it has a subjective answer.)

    my argument is that tennis is a human sport so it should be based on an observation and judgement made at an objective human level.

    hawk-eye defeats the purpose of the game. it's like looking at tennis through a microscope.

    hawk-eye in tennis is like tracking which blades of grass are white and which are green in football/soccer/rugby for the purpose of determining if a player is out-of-bounds by zooming into each footstep at a magnification of 1000x to see if a single white blade of grass may or may not have touched his foot. that would be ludicrous, so is hawk-eye in tennis.

    or it would be like if i asked you what's the answer to:

    1.234 * 5.678

    and you told me:

    7.00665200000000000000

    when anything beyond 7.007 is irrelevant. that is hawk-eye in tennis.
     
    #41
  42. woodrow1029

    woodrow1029 Guest

    If its margin of error was 4 mm, it would never be used in professional tennis.
     
    #42
  43. Magnetite

    Magnetite Professional

    Joined:
    Jul 7, 2011
    Messages:
    907
    This would work well, although unlike what others have said, you would still have to keep lines people in case their was a malfunction.
     
    #43
  44. nousername

    nousername Rookie

    Joined:
    Sep 26, 2007
    Messages:
    317
    an umpire would make terrible calls if he had 12 camera angles at 1000 frames/second at his disposal????
     
    #44
  45. Magnetite

    Magnetite Professional

    Joined:
    Jul 7, 2011
    Messages:
    907
    Yes.

    He's human. Under pressure situations, he may get stressed and rush his decision. It happens all the time, and would take too long anyway.
     
    #45
  46. pennc94

    pennc94 Semi-Pro

    Joined:
    Jul 26, 2006
    Messages:
    759
    Location:
    Maryland
    I am not sure you understand what I wrote.

    I am FOR the Hawkeye or any computer line calling system (because it appears to be an improvement over the human eye)
    I am NOT FOR the challenge system (because it keeps players from focusing on just tennis).

    However, since Hawkeye has some measure of inaccuracy (as small as it may be) it should be addressed (unless players are willing to accept the stated margin of error).

    If the line-calling by Hawkeye is instantaneous, then linesmen should be eliminated and have Hawkeye make all calls. Keep the umpire for other rule enforcement.
     
    #46
  47. j00dypoo

    j00dypoo Rookie

    Joined:
    Dec 22, 2010
    Messages:
    275
    Location:
    Atlanta
    well of course tennis is a human sport - robots don't play it (well maybe in japan but I haven't seen it yet). Those two points have absolutely no relation to each other.

    Cars are purely mechanical machines, so only mechanical machines should operate them.

    Also, even if the average margin of error were 3.6 mm (where did you find this btw), that's such a small margin and much better than any human could ever hope to achieve. Seriously, take out a ruler and measure out 3.6 mm. How could anyone, even using high speed cameras, hope to get a better measurement and judgement call in as short a time as hawkeye is able to produce?
     
    #47
  48. nousername

    nousername Rookie

    Joined:
    Sep 26, 2007
    Messages:
    317
    perhaps, but i doubt it.

    to me, the one argument for hawk-eye is the speed of decision. it does comeback quick with an answer. but with the proper setup, e.g. for a trivial cost (in the grand scheme of an ATP event) adding monitors to all umps' chairs would make video reviews quick.
     
    #48
  49. Lsmkenpo

    Lsmkenpo Hall of Fame

    Joined:
    Aug 21, 2006
    Messages:
    3,915
    From the link to the article I posted earlier in the thread, Paul Hawkins, managing director of Hawk-Eye technology states hawkeye's average margin of error is 3.6mm.

    The ITF rulebook also states an electronic line calling device can be used as long as it is accurate within 10mm.
     
    Last edited: Dec 13, 2011
    #49
  50. woodrow1029

    woodrow1029 Guest

    That article is from quite a while ago. Since then, the technology has improved and the margin of error is around 1-2 mm.

    The ITF rulebook does not state that at all.
     
    #50

Share This Page