Forgot your password?
typodupeerror
AI Robotics Hardware

Go For It On Fourth Down? Ask Coach Watson 241

Posted by Soulskill
from the falling-back-on-the-old-destroy-all-humans-play dept.
jbrodkin writes "If humans can't beat a computer at 'Jeopardy!' why should we trust them to make the right call on fourth down in the Super Bowl? That was the fundamental question asked by some researchers at the recent MIT Sloan Sports Analytics Conference. With thousands of variables to consider on the basketball court or other fields of play, it only makes sense to let computers handle questions of strategy, says Tarek Kamil, whose company built a chip-containing basketball which takes 6,000 measurements per second. 'Fifty years from now, we're going to laugh about how we used to give coaches this much responsibility,' he says."
This discussion has been archived. No new comments can be posted.

Go For It On Fourth Down? Ask Coach Watson

Comments Filter:
  • But the Super Bowl and Fourth down are football things. Not basketball things.

    As for sports at the upper levels, there is more involved than merely picking the correct play. You need not only the play, but the execution of it. Coaches do far more than just come up with strategy, they also, as the name implies, COACH.
    • by blair1q (305137)

      I bet you'd do as well on the field if you had a computer picking your plays and feeding Duke Nukem quips into your helmet radio.

    • by antdude (79039)

      I thought this football thing was about soccer. [grin]

    • by sconeu (64226)

      Can someone put it in NASCAR or F1 terms? That way we'll have a car analogy, and everyone will be happy (except for the Indy car guys).

  • by Haedrian (1676506) on Monday March 07, 2011 @05:00PM (#35410938)

    Why should we even let humans obey the orders? Machines can do it more efficienty.

    And then why do we need to do it in the physical world? It might be more interesting if there's no gravity, or higher gravity or something.

    So the entertainment of the future will involve us seeing computers play video games in front of us.

    • by TrippTDF (513419)
      Eventually, we will just analyze everything that makes us enjoy a sporting event, and the computer will just spit out a game that we will enjoy watching, for sheer entertainment value.
      • by Abstrackt (609015)

        Eventually, we will just analyze everything that makes us enjoy a sporting event, and the computer will just spit out a game that we will enjoy watching, for sheer entertainment value.

        I hope it's per user rather than what the majority votes for, otherwise 95% of the games will be a bunch of guys getting kicked in the balls.

      • by Surt (22457)

        The real challenge will be to do it so that we can bet on the game in vegas.

      • by corbettw (214229)

        As long as it involves a shower scene with Cameron Diaz, I say go for it.

      • by swb (14022)

        Couldn't they come close to doing this now?

        Ask the user what kind of game they enjoy -- ie, for baseball -- low-scoring pitcher's duel, high scoring home run fest, lots of good defense, etc.

        For data inputs, use the accumulated statistics for teams and players they already have. Build composite teams of players who are statistically most likely to provide the game the user wants to watch. Factor in stadiums, seasons, weather, etc.

        Have those two teams "play" and use an existing 3D game engine to display the

      • by guruevi (827432)

        With computer games coming pretty darn close to realistic it's only a matter of time before we can simulate full games without people seeing the difference. Most high-profile games are already fully digitalized with lines and arrows being drawn on the field in real-time.

        The thing that makes sports enjoyable to watch however will have to be simulated as well - the sheer unpredictability of humans, rapid and unexpected changes in strategy in response to the other side's lineup, injuries etc.

    • by dwandy (907337)
      Because sport is about human competition. There is a distinct difference between having a machine judge an outcome, or having a machine aid with planning and strategy and having a machine perform the athletic task.
      Other sports (like automobile racing) have been deciding where the line is between "computer-aided" and "computer-performed" for some time now; it's only natural that other sports begin to grapple with this problem.
      Ultimately there will always be a market for those that want to see how we compet
    • I only watch sports because it's great to watch humans push the limits of physical ability! The thought of watching machines is boring.
    • And since a computer's actions are deterministic, time-shifting them really won't affect anything, since we can know the outcome in advance. As a result, sports will become little more than virtual actors in a film produced purely for our entertainment, rather than out of a love for the spirit of what once made it great. In that regard, the future is here now, since George Lucas has already started using robots to replace actors in his films. ...

      What do you mean Hayden Christiansen is a real person?

  • Jeopardy is not Strategy. Jeopardy is taxing the computers ability to understand reasonably complex language, and find the solution to a question. A computer has a database of answers, and there is only 1 correct one to select. Football strategy does not imply only 1 right answer. It relies on so many more things than that. All in all, a pretty ridiculous statement.
    • by Haedrian (1676506)

      Actually if you break the environment into a large amount of variables, then its a textbook example of how AI would work.

      That said, the bonus of a computer is how its able to process many things quickly, and not that it will learn faster than humans (by examples). So I'd say that if you get a coach to act in 50 games or so, he will be able to give better results than a computer who acts in 50 games. You'd need a large amount of games in order to get a difference. Now a computer could easily process thousand

      • by hedwards (940851)

        It's a lot more complicated than that. Sure you can break it down to variables, but what exactly does that get you? How do you calculate like likelihood of a dropped pass when neither team is used to playing in the cold? Or the likelihood of a QB being sacked as a result of environmental conditions when considering that aspect of blitzing?

        Plus, you're not going to see a computer calling an onside kick to start the first half like in that Packers game last season against the Patriots. A lot of coaching has t

    • by pz (113803)

      Answer: "attempt field goal with 10 seconds remaining in the 4th quarter on the 25 yard line with 4th down and 3 to go when down by 2 points"

      Question: "what is a winning strategy?"

      Seems like that fits the description of only one correct answer to select. It all depends on how you phrase the inputs and how efficient your search is.

      The point is that the success of the Watson team is twofold, first, as you rightly point out, to understand complex human language. The second is to efficiently organize and sear

      • by umghhh (965931)
        While we have machines getting better at understanding the gibberish humans produce we may try to get a machine to convince people to get over the gibberish in the first place. This could involve significant simplifications to our speech patterns down to say: 'yes, sir', 'no, sir', 'I do not know sir' and "I have not understood the order sir'. If we are not allowed to respond in any other way this can increase meaningful content in an average speech sample.
      • by Panaflex (13191)

        Clearly I have defeated this earthworm with my words – imagine what I would have done with my fire breathing first!

    • Jeopardy is not Strategy.

      Gotta disagree here. Apparently the big reason Ken Jennings was able to stay on top (aside from his recall ability) was the fact that he knew exactly when to press the buzzer. If you press before Trebek finishes the question then your buzzer is disabled for a short period, so if you know exactly when to press then your more likely to increase your chances to answer.

  • It actually worries me a bit that we're letting computers make decisions for us. I can see doing labor and computations, but when you place the machine in the decision making process and the human follows, then the machine has become the master. I know right now it's all well and good and very acedemic, but I really REALLY don't like the idea of say, a machine manager. People feel that we're already cogs in a great machine, this is just a baby step towards a very scary future.

    Also, when we let machines make

    • by blair1q (305137)

      We as a people stopped growing intellectually at about the time we started allowing money to make our decisions for us.

    • You are not the only one thinking on those lines [marshallbrain.com].
    • by Bucc5062 (856482)

      It could be based on perspective. Asimov wrote a couple of short stories with positronic brains making world decisions. [wikimedia.org]

      Given the nature of the human condition these days, perhaps the overused meme "I for one welcome our life decision making robots" is better then the current option, greedy, political, me oriented humans.

    • by blair1q (305137)

      that came true the day the clock was invented

  • Replace the coaches and players with robots. Or just simulate the whole thing in cyberspace.

    Games are exercise. Pro sports forget that. A big lot.

  • by kenrblan (1388237) on Monday March 07, 2011 @05:05PM (#35411018)
    If both teams relied solely on computer models to make the decision, both teams would likely know whether an attempt on 4th down would be attempted. There would almost never be an unexpected attempt, and the only unaccounted variable would be the actual play to be run on the attempt, which could also be predicted relatively accurately by considering coach play calling tendencies.
    • until someone writes the script

      `/dev/rand | chooseplay`

    • by MobyDisk (75490)

      The only way to win is not to play.

      • by blair1q (305137)

        The only way to win is not to play.

        Just don't let your enemy know that's your strategy.

    • by toastar (573882)

      ... and the only unaccounted variable would be the actual play to be run on the attempt, which could also be predicted relatively accurately by considering coach play calling tendencies.

      They don't call it game theory for nothing.

    • All I heard was: "You've beaten my giant, which means you're exceptionally strong, so you could've put the poison in your own goblet, trusting on your strength to save you, so I can clearly not choose the wine in front of you. But, you've also bested my Spaniard, which means you must have studied, and in studying you must have learned that man is mortal, so you would have put the poison as far from yourself as possible, so I can clearly not choose the wine in front of me."

    • Now, a clever man would put the poison into his own goblet because he would know that only a great fool would reach for what he was given. I am not a great fool so I can clearly not choose the wine in front of you. But you must have known I was not a great fool... you would have counted on it... so I can clearly not choose the wine in front of me!
    • by pz (113803)

      If both teams relied solely on computer models to make the decision, both teams would likely know whether an attempt on 4th down would be attempted. There would almost never be an unexpected attempt, and the only unaccounted variable would be the actual play to be run on the attempt, which could also be predicted relatively accurately by considering coach play calling tendencies.

      I suppose you're claiming that a model could never be built that had the value of surprise or unusual choices as one of the variables. And a random element is right out.

      More seriously, if you can think of general characteristics of anything, it's usually pretty easy to come up with a model to cover it. Especially when big money is involved. The same reasoning posited above would argue that Watson was impossible to build. I'm glad the very talented folks at IBM did not succumb to such shortsightedness.

  • there are two risk reward ratios

    the ghetto risk where you risk a lot for little potential reward. choose any stupid scam you read about where the idiot criminal gets caught for stealing very little money in the big picture. say a few thousand $$$.

    the good risk where you risk a little or even a lot but for a good reward. like say finding a new job at google.

    fourth down is a bad risk unless you're losing and there is very little time left since most plays in sports result in very little reward or none at all.

    • by ShadyG (197269)
      Depends. Some teams have less than stellar placekickers. If you can't reliably make a field goal from the 30 yard line, and you have 4th and 2 at your opponent's 35, maybe you go for it mid-game. Any punt made from that position can only potentially help you out in field position by 34+ yards, and if you try you run the risk of the ball going through the end-zone netting you all of 15. Add to that a missed field goal turns the ball over at the spot of the kick, not the line of scrimmage, so you'd be set bac
  • Seriously, there was a Simpson's episode [wikipedia.org] about this very topic.
  • My football has only three downs [www.cfl.ca] you insensitive clod!

    Actually, the blog 55 Yard Line has an excellent article on whether or not more coaches should go for it on third down [yahoo.com] based on yards to go for a first down and the line of scrimmage.

    • by blair1q (305137)

      Someone ran the numbers for the NFL about a decade ago. I've been carrying it around in my PDA/Palm phone/Google phone ever since.

      Coaches, to a man, vastly underestimate the value of going for it on 4th down. Vastly.

      • by Raenex (947668)

        Coaches, to a man, vastly underestimate the value of going for it on 4th down. Vastly.

        You might recall all the heat Belichick got for going for it on 4th down instead of punting against the Colts.

    • by Locke2005 (849178)
      Frickin' Canadian wimps can't handle four downs like real men! The only thing you Canucks do as well as Amerkins is pat each other on the butt after a play...
  • How do we quantify a human characteristic like courage? Knowing a player is competing despite having a minor injury is one thing; knowing how hard he will compete and how well he will play is something else. Until we can quantify both qualitative human characteristics like courage, fear, ambition, and stubbornness, as well as the "gut call" a coach makes based on his impressions of the individuals on the field and how they function as a group, computers aren't going to be better at this. They will proba
    • by MobyDisk (75490)

      Until we can quantify both qualitative human characteristics like courage, fear, ambition, and stubbornness, as well as the "gut call" a coach makes based on his impressions of the individuals on the field and how they function as a group, computers aren't going to be better at this

      We sure can. It is just statistics.

      You have a 27 year old black running back with a left ankle injury sustained 5.6 days ago. He weighs 275 lbs, has 2 super bowl rings, is right-handed, with 2 felony counts of rape (1 acquitted, 1 pending), who has run 67 yards this game with is 4.3% below his average. He is against a team with a 6-3-0 WLT record. He plays 12% better when his team is behind, runs 14% father in the red zone, and 3% worse when his mother-in-law is in the stadium. He plays 1.7% better dur

  • I don't care about strategy, but the NHL is lousy at telling if the puck is above the height of the crossbar when a goal is deflected out of the air. My team has been obviously burned on this twice this season, one game-winning goal for the other guys allowed and another disallowed for us that should have counted. I would love for stats so the ref could instantly tell if a puck crossed the goal line under a goalie, too.
  • If humans can't beat a computer at 'Jeopardy!' why should we trust them to make the right call on fourth down in the Super Bowl?

    Good point. I will replace the head coach of my NFL team with a computer as soon as I'm done having dinner with the queen of England.
  • You're an editor at some network-central industry rag. One of your staff writers heads up to Boston to attend the Sports Analytics Conference. He rubs elbows with big names in professional sports, attends a few break-out sessions that discuss gathering data to reduce player injuries and data acquisition inside basketballs, and then writes up a rambling article to justify his expense account. What's are you to do? Still hung over from the "Watson vs. the humans" Jeopardy! party, you write the headline: "Go f

  • The example given was a tactical, not strategic decision.

    "For the honor of the regiment" /obscure

  • Just simulate it and publish the score.

    • by Haedrian (1676506)

      Also simulate some 'special moments' that the news can show on highlights, grab a player at random and give him unwarranted self-importance, then simulate him having a controversial life so people would have stuff to gossip on.

      I'm sure most people won't notice the difference.

  • by Surt (22457) on Monday March 07, 2011 @05:22PM (#35411260) Homepage Journal

    The robots will be laughing at the robot faction that claims there were once biological beings on the planet.

  • "Fifty years from now, we're going to laugh about how we used to give coaches this much responsibility"

    WTF? Talk about completely missing the point.

     

    • by blair1q (305137)

      Bullseye.

      This is why we desperately need to stop giving pro sports leagues monopolies.

      This stuff needs to be diluted down to its rational level, and fast.

      • This is why we desperately need to stop giving pro sports leagues monopolies.

        "Monopolies"? There's competition: football, baseball, basketball...

        And, of course, you are free to start your own football league (or foosball if you prefer).

        This stuff needs to be diluted down to its rational level, and fast.

        "Rational"? It's entertainment. I don't find it interesting myself but if others want to amuse themselves with it why should I care? It's their money, not mine.

  • Fifty years from now, we're going to laugh about how we used to give coaches this much responsibility

    I hope not; coaches are essential to the nature of a team; these are human competitions, and if we're not considering robotic players, we have no business looking into robotic coaches. Each coach has his or her own take on what should happen, including intuition, foresight, insight, and motivational talking. No computer is going to be able to give an exhausted player a second wind by talking about probabilities and describing an unrehearsed play.

    Consider American football, which is often dubbed a chess

    • by gknoy (899301)

      However, having access to the odds of a play's success would be very useful to those coaches. Imagine two poker players, and one knows more than the other about the odds of particular hands, or the likelihood of the next card being something he needs, and over time games should work out in his favor.

  • Because humans are fallible and can't comprehend as many variables as a computer. It's that lack of perfection that makes games great.

  • First thing - computers can't beat humans at strategy games where there are huge variables in risk versus reward, such as No Limit Texas Hold em. They can use strategies of course but a pro player can tear them to pieces. See the notes on Man versus machine where Polaris went up against poker geniuses like Bryce Paradis and then read his post analysis if possible. And that was a game of limit. No limit adds a layer of complexity that makes the mind boggle. And that's a simple game. Imagine applying strategy

    • by blair1q (305137)

      Computers routinely kick humans' asses at No-Limit Holdem.

      • by gknoy (899301)

        Are they beating skilled players, though, or people as bad at poker as I am? :) I imagine that beating scrubs is comparatively easy, even if beating masters is hard.

      • Unless you have some proof of winning at skilled levels I advise you "ship it". Writing a bot to work scrubs using .05/.10 using pokertracker's using VP/PR/AF and auto hotkey is not winning against a skilled player. When one of the world's most powerful machines can't beat a field of high _limit_ players then where is the one that beats high stakes no limit players? It doesn't exist sir.

  • And I thought that the reason that statisticians weren't good as sports coaches or generals was because there were 1000's of variables that couldn't be quantified or computed.

    oh... wait...

    • by blair1q (305137)

      Pretty much the case. Each game depends on a large number of probabilities, so the number of outcomes is nearly infinite.

      About half of what we attribute to skill, especially when it comes to in-game coaching, is luck. The skill is in the management of the misperceptions that lucky events impute to the person making the decisions.

      It would take a computer trained in psychology and criminal psychopathy to be Bill Belichick.

  • Don't computer games, Madden and others, basically already do this? At least on a simplistic level?

  • by travdaddy (527149) <{gro.liamxunil} {ta} {ovart}> on Monday March 07, 2011 @05:38PM (#35411494)
    Any coaches not already using computers to help with their strategy are doing themselves a huge disservice. I'm sure they are already crunching all the statistics they have, like how often a 4th down conversion works and using that to help with strategy.

    However, I'm not sure this microchip does anything that anybody is interested in. It probably costs a fortune too, and they're putting it in a ball that's going to get knocked around? I'm sure the NBA is not sold on paying a lot of money to find out, as the article mentions, whether Johnny is 14% more dominant with his right hand than his left. And useful stats like "Time of Possession" will still have to be done by a human.

    In fact, I can't think of a single stat important enough for a microchip in a ball to transmit in real time, and even if it was, it transmits to BOTH coaches. It's in the ball, so it creates somewhat of an arms race and just creates more information and work for the coaches to consider.
  • by Sir_Sri (199544) on Monday March 07, 2011 @05:39PM (#35411516)

    I research AI for strategy.

    A computer is really good at finding optimal strategies if you can properly quantify relevant variables. In sport there is advantage in not taking an optimal strategy, because your opponent won't know which non optimal strategy you've chosen until it's too late. If you're going to use randomness to determine which strategy to use, then the computer is no better than a coach.

    That assumes, probably wrongly, that you can quantify what's going on. Is that opposing quarterback's limp important, a fake, how serious is it (numerically)? Even if a computer is good at predicting one particular game, (say the superbowl) that would be based on the data from the whole of the rest of the season to assess how good the players are.

    There's a lot of sport to be had in running AI's against each other, especially based on the same sets of data and see what they do. But that is a *very* different problem from actually simulating a real match, yes, the average of 10000 trials may be correct, but there are only a few real games, not thousands. That randomness, sportmanship, and people doing extraordinary, unexpected and great (or stupid) things is what separates a real match from a statistical model.

  • Ocean's cluster of ninety IBM Power 750 servers (plus additional I/O, network and cluster controller nodes in 10 racks) with a total of 2880 POWER7 processor cores and 16 Terabytes of RAM. Coming soon to a theater near you!!

    Can the two break the banks at major Las Vegas casinos? The difficult part is getting Watson to do all those spectacular acrobatic tricks. I guess they'll have to use a stunt Watson. Or maybe Watson can generate his own stunt scenes, just like Jackie Chan does his own stunts?

    Um, s

"Just think of a computer as hardware you can program." -- Nigel de la Tierre

Working...