Slashdot Mirror
Game Theory at 190mph
cameronm writes "A recent article in Slate discusses the value of NASCAR racing as a tool to study Game Theory. You can view the original study at FirstMonday."
← Back to Stories (view on slashdot.org)
← Back to Stories (view on slashdot.org)
It is a branch of mathmatics.
4 86 251012.html
http://shop.store.yahoo.com/doverpublications/0
KFG
From what I've learned through discussions with programming mentors, and analysts, Game Theory is the study of a game's impact on the mental processes of a human mind. It stems from the constant 'Violent games make our kids violent', looking for proof to (dis)prove that statement. It focuses on what-if's like working in this situation under pressure, with x amount of experience (hours in a cockpit), tools available, how long will someone wander around in a maze before freaking out, tendancies (driving a car in Grand Theft Auto 3 into a storefront as opposed to doing so in the real world). Do people take more risks in a simulation than they do in real life? Is it more effective?
Y'know, that sort of thing.
Informatus Technologicus
Game Theory (GT) is a mathematical field all about choices: given a set of knowledge, build a set of choices that will lead to the greatest expected outcome.
GT can be applied to games, since games generally consist of a set of choices. But the term is more often heard in economic circles: stock market, insurance calculations, portfolio planning... But don't limit it to that either. GT can be applied to most any sort of competition, such as competition for food sources between predator species, or trying to find the shortest/quickest way through the lines at a supermarket.
A number of different factors influence your choices. How much information do I know? How much information does my opponent(s) know? Do they know I know what they know? Can I influence my opponent's choices such that we both get a better outcome? All these things will change my strategy.
Well, if you don't win, you don't get paid. As it should be. Most drivers, say Dale Jr, Michael Waltrip, Dale Jarett, Tony Stewart and Jeff Gordon win a few times a season and barring winning, finish in the top ten. Plus if you get high in the points race you get paid money too. So saying that they don't get paid millions is not showing much insite there.
On the other hand, most teams and drivers will bend over backwards to do things for the fans. They have appearances, they sign autograph after autograph at special sessions and right in their locker room, the garage area. This has slowed down a bit. It had gotten so bad that they had to issue Hot and Cold passes because asshole fans got in the way (last year, Tony allegedly pushed a fan out of his way on the way to his trailer. Ends up they were just trying to egg him on.). The new rule was made to protect fans from flying tools, cars, and to give the racers some time to get away. Now they can walk from their RV to their trailer without a crowd behind them when it's close to practice times and race time. Imagine if football or baseball was this way even a little. It would never happen.
Very rarely do you hear racers bad mouthing each other and 2 years ago when Dale Earnhardt Sr. Died, there was an outpouring from fans AND drivers.
NASCAR drivers are usually clean cut. OH they may toss back some clydesdales, but that's about it. You don't hear much about this driver is doign this and this driver is doing that. Noone to my knowledge has ever been caught doing drugs and you usually don't hear of them beating their wives either. They are generally alot better then the majority of the public, but human all the same.
NASCAR is more then just getting in your car and going fast. You got to MAKE SURE your car is fast. Tweak the camber hear, round of track bar there...a few tenths of a pound of air in a tire, patching your car so it's still aerodynamic. NASCAR racing can be incredibly geeky and usually is. Races like Daytona and Talledega are like chess matches instead of races. Get the help of the draft to get ya up front. Block the others to stay up front. No when to pit and when to stay out. When to two tire and when to get 4 new tires.....on and on. Crew Chiefs not only have to know alot about cars, they have to be able to interpret the driver and tell the pit crew what to do on a pitstop. They calculate fuel mileage so they know to the lap when they can stop to get tires and gas. They also do quick thinking on the spot when a driver has some damage to his car. NASCAR is one of the most complex sports of all since it's really more then just the driver and the car. Recently, they added crew names to the pit crews uniforms. Without men like Chocolate Meyers and Slugger Labbe, these guys would be driving Go Karts. NASCAR IS A COMPLEX sport as well as a honerable one. One I am proud to show my son and say go ahead and give it a try Not like other sports where you don't need people of high intelligence.
Gorkman
Not to mention the fact that the pressure on them during a pit stop is immense. Competition is so tight that a second or less in the pits could make the difference in winning the race.
Here, read this. Game theory is the attempt to take common situations in games, and apply them to sociological situation. THe idea is that most human interactions are at some level either simple or complicated games.
Game theory does things like take the idea of, if you are one of three people in a three-way duel, and you hit your mark 20% of the time, and you are up against two opponents who one hits their mark 40% of the time and the other hits 60%, who should you aim at? And the answer is to just fire your gun into the air, becuase of course in that case the chances are best that the other two will just kill the other off.
Game theory tries to take scenarios like that and use it as a metaphor to describe the best course of action in, say, warfare, where you are a weak power and you have two enemies that are strong powers. The assumption is that people will not act all that differently when engaging in transnational diplomacy than they will when playing poker, if the situations can be said to be comarable.
I am not a big racing fan and have never really understood the allure, but the section Basic Dynamics of Drafting is a fantastic read. It gives great insight into tactics used by these highly skilled drivers.
The original study has implications for erstwhile competing entities that are forced to co-operate to suceed. Comparison is made to the tech sector. RTFA.
Cooperation in competition to win and GT? Check out Tit-for-Tat, as well as a bunch of other things for more examples.
My favorite part was: "It takes two to pass one.""There is no teacher but the enemy."-Mazer Rackham
John Nash is _not_ the originator of Game Theory. John von Neumann is. Do a google search on him.
Nash had many interesting ideas relating to all sorts of fields including economics and game theory, but he did not originate either one. Von Neumann, on the other hand, created an entirely new field of mathematics which is interrelated with economics, political science, sociology and others. If you ever get a chance to read von Neumann's books, I highly recommend it.
---
"Of course, that's just my opinion. I could be wrong." --Dennis Miller
European racing grew out of an entirely different concept. There the idea was the "test." Pitting the machines of different manufacturers against each other to see which one could best negotiate its way between two points over the road. The driver was considered largely incidental. There merely because someone had to be there to operate the machine.
This is incorrect. While european auto racing did stem from testing the superior vehicles that europe produces, the driver has always been a celebrated element. The "checkered flag" which nascar fans seem to enjoy so much was first used by the Germans.
My largest (technology oriented) complaints with nascar are the ancient technology used in the engines (pushrod V8's, what is this the 1950's?) and the fact that it's called "stock car" racing. There is not one single part in common between the nascar ford taurus and the ford taurus you get at the ford dealer. nascar is not stock car racing. Real stock car racing is called "Touring Car" racing, where the cars are basically showroom stock, as they are at any dealership.
I'd rather be a conservative nutjob than a liberal with no nuts and no job.
Drafting is the central feature of bicycle racing, and in a different form temporary cooperation is common in running (though less for drafting and more to prevent yourself from getting jostled). It also happens in adventure racing, which is popular enough that its biggest event (the Eco-Challenge) is fairly well-known.
demi
I suspect NASCAR is not an ideal (ie: boundary pushing) example of the sorts of game theory this article talks about.
Most of all: cycling. There's more flexibility for overtaking - a limitation in track racing. Other than that, the nature of the competition is similar - slipstreaming and darfting.
Another comparison I thought of is the board game Diplomacy, because there's more time to think, and betrayal is all but inevitable: in order to win, you will have to screw your allies if you are on the path to success. This is not necesarily the case in nascar where you may be happy to lose now because it's meaningless whether you come fourteenth or fifteenth. In diplomacy, there is a status attached to mere survival. I admit, there is a path to stalemate whereby you honour your agreements. However, it is rare.
However, I found the point of the article - regarding where accidents happen - to be very interesting.
Believe with me, my saplings.
If you want to see cars which bear some resemblance to production race, it's either a production car category like Porsche Cup, or group N rally (the World Rally cars are very heavily modified).
My favourite racing categories are actually motocross bikes or dirt speedway racing, which in terms of spectator entertainment crap all over the open-wheel categories, NASCAR, Touring Cars, and even rallies (you don't get to see enough of the course in a rally, sadly). The other nice thing about supercross is that you can actually buy the bikes the pros use (well, not quite, but very, very close).
Any sufficiently advanced technology is indistinguishable from a rigged demo
--Andy Finkel (J. Klass?)
"Most drivers, say Dale Jr, Michael Waltrip, Dale Jarett, Tony Stewart and Jeff Gordon win a few times a season"
Actually, most drivers are lucky if they win a few races in their entire career. see for yourself
204mph civic. That was with two seconds of googling.
autopr0n is like, down and stuff.
You should check out The History of NASCAR...very interesting read...many ppl know about the moonshine thing, but how many know about the real history of the sport???
There's a lot of science and engineering involved with NASCAR...which is why the modern drivers are more engineers than mechanics...
This is also one reason why I think the current drivers and crews are much better role models than most of our athletes in traditional sports (football, baseball, boxing, etc)...
who enjoys a good stock car race (or, more accurately, I like watching darned near anything race...) I'd like to add in a few notes:
1) Stock car racing isn't always like this; this is the norm at Daytona or Talladega, but smaller tracks have different dynamics. Also, simply saying this is "NASCAR" is also misleading; the "NASCAR Craftsman Truck Series" is just as much NASCAR as the "NASCAR Winston Cup Series" they're talking about here, and the trucks run differently, even at Daytona. (The trucks have unrestricted motors, and instead rely on the fact that they have to punch a bigger hole in the air to keep them at "safe" speeds. This leaves sufficient power to bring back one-on-one moves like a slingshot)
2) NASCAR drivers aren't all "he". Shawna Robinson, Deborah Renshaw, and Tina Gordon would probably argue that point.
3) The comment that racers get more aggressive when they're worried about losing more so than winning is questionable-- it seems more to me that the agressiveness level is a function of how many laps are left, and not position on the track. The reason backmarkers tend to wreck more often is their car isn't handling as well, which is why they're back there in the first place...
On the other hand, the game theory aspect is pretty spot-on, and it gets even better than what the article noted: Many race teams field more than one car. So there are some cars out there that a driver can trust more so than the others, since they're teammates. Finishing second to your teammate isn't nearly as painful as finishing second to somebody else-- pays the same, but if you didn't win, it's much better to have not won by helping your teammate do so. The game mechanics are notably more complex than the article notes, and may even be as complex as the auto mechanics...
...I was reading about this exact same subject, as it related to bicycle racing at least 5 years ago, if not more. And it's for the same reasons -- two competitors will need to take turns drafting off one another to get to the finish line before the peleton.
To make matters more complex, those two racers have to have enough guile to draft longer than they pull...so that they have more energy for the sprint against each other for the win.
Still...that racing where drafting is involved (motorsports, cycling, whatever) is extremely complex from a game theory perspective is nothing new...
blog |
"sea of Confederate flags"
I have been to many NASCAR races (both up North and down South) and I have never seen anyone waving a Confederate flag. I *think* I may have seen a sticker of a Confederate flag on a window in TN at a NASCAR event.
NASCAR fans are NOT all southern rednecks. It is a mainstream sport with fans of all races, sexes, and monetary backgrounds.
Holy s-, it's Jesus!