If basketball were as timeless as chess, old games and old analyses would still have meaning and interest today. The NBA would make a fortune rebroadcasting old games. There is a way to do it: shine bright lights on the NBA's most closely guarded secrets.
Philip Maymin
Basketball News Services
Have you ever recorded a basketball game intending to watch it later, only to find out the score and delete the game unwatched? The contest loses most of its suspense when you find out who the eventual winner is. But does it have to?
Strategic games like chess can be analyzed and appreciated decades after they've been played, even when the winner is common knowledge. There are columns in newspapers analyzing a single chess position or bridge hand, but no chalkboard analyzing a particularly brilliant play or possession.
How can we make basketball timeless?
Should it be timeless? Perhaps it is no different than flipping a coin. What do I mean by that? I mean two things: first, if you know that a particular coin toss is going to end up heads, you don't care how it's flipped or by whom or how the people were standing. It's just a coin toss! And second, even if you don't know how a particular coin toss is going to work out, you don't care much how the player tosses it or if he rubs it for luck or anything else. It might make good drama, temporarily building up suspense, but in the end your gut tells you none of those lucky charms really matter. It's just a coin toss!
A coin toss can never be timeless. It is a temporary illusion of drama, like slot machines or video poker. You can feel that you're about to get "hot" or that you need to press the buttons in a certain order to enhance your chances of success, just like in basketball you can wear lucky shorts or socks or shoes, but in the final analysis, you know it's just luck. You're never going to watch old slot machine rolls on game tape, trying to analyze strategies and tendencies.
Is basketball just a coin toss? I fell asleep before the Timberwolves-Lakers game finished last night, and I watched it this morning. If you've done the same and don't know the score yet, I won't spoil it for you. It took every ounce of laziness on my part to not find out the final score before watching the tape. I know if I had, the game would have lost a lot of its interest. Does that mean basketball as a sport is just a coin toss?
In my heart I don't believe that's true. This is a game I love and analyze and think about constantly: I don't care a whit about slot machines. I don't lie awake at night thinking about better strategies for picking lottery numbers. But there seems to be something about basketball that is related to skill and not luck.
I could make a case that basketball is mainly luck, and it would go something like this: most games are decided by five points or less. Free throw shooters shoot about 50 percent from the line (an assumption for simplicity). There are about 20 free throw attempts per team per game. What are the mathematical chances that two random samples of 20 coin tosses differ in the number of heads by more than five? We can calculate that like this. Call the two samples A and B, with A being the one that gets more heads. If A gets all 20 heads, B must have 1, 2, 3, ..., or 15 heads, but not more. If B gets 19 heads, B must have 1, 2, 3, ..., or 14 heads, but not more. And so on. The probability of A getting all 20 heads is 50 percent raised to the power of 20, a ridiculously small number. The probability of B getting 15 heads or fewer is the sum of the probabilities of getting exactly one head, or two heads, ...., or 15 heads. Both of these numbers can be calculated easily in Excel, using the BINOMDIST function. You then multiply those two probabilities to get the probability of A having exactly 20 and B having no more than 15. Then you do the same thing for A having exactly 19 and B having no more than 14. And so on down to A having exactly 5 and B having exactly zero. You add up all those multiplied probabilities, and then you double that sum. Why do you double it? Because you don't know which sample will be A (the one with more heads) and which will be B (the one with fewer heads). The answer comes out to 15 percent.
That means that based purely on luck and free throws, you would expect only about 15 percent of games to differ by a score of more than five points.
And that's pretty much what we see. About one in six or seven games is a blowout but the rest are pretty close. (As an aside, the appropriate score to look at is probably the one with about a minute or so left in the game, because oftentimes a four-point lead can become a seven- or eight-point lead as one team starts fouling the other to stop the clock.)
That would be the case I would make that basketball is mainly luck. I could tinker with the formula, making the probability of hitting a free throw 70 percent and the number of free throw attempts per team per game 30, and that would increase the probability of a blowout to 20 percent, meaning you'd expect only four out of five games to be tightly-contested.
In the end, though, my analysis would be severely flawed, and here's how: what looks random need not necessarily be random. There are about 100 possessions per team in every game, and the winner is the one who has the better average. The difference between two averages of two distributions is almost surely to be statistically significant after 100 trials. The statistical Law of Large Numbers starts to apply: as the number of data points you're averaging over increases, the distribution of the results appears more and more like the normal distribution. And the more it looks like the normal distribution, the easier it is to statistically discern winners from losers.
Despite all that analysis, I don't think it's mainly luck. Or at least, I refuse to believe that it's merely luck. Every game I've ever played, sure there have been times we've won or lost despite the odds -- I don't disagree that luck plays a role -- but on average you tend to beat worse teams and you tend to lose to better teams. And you can tell pretty quickly who the better team is.
The interesting thing is it's not always talent that differentiates: it's the plays they run and their teamwork and their ability to know where the other guys are at all times. If you play with people long enough, you know where they are likely to be, and you can almost make blind passes to the right spot every time.
That's what I'd like to see analyzed! That's what should be in every newspaper next to the bridge and the chess problems. Here is the Lakers-Spurs game five, Lakers have the ball with 0.4 seconds on the clock. Here is the initial configuration of how every player is standing on the court and where the inbounds pass is coming from. Lakers to play and win.
That kind of analysis of particular plays makes it interesting to watch past games. Most historic free throw attempts can just be fast-forwarded through, and most possessions with turnovers are just unforced errors, like a mistake in a game of chess, typically marked with a question mark in the log.
In analyzing chess games, many moves are standard moves, especially in the beginning, that you skip through. You focus on the key positions and where a new plan developed.
In basketball, you want to focus on the first few possessions that a coach made a defensive switch. Should Flip Saunders have Kevin Garnett defend Kobe Bryant? Here's how the first few plays played out. See how Kobe does this or KG does that. That's interesting!
Look at this play that the Lakers run all the time in their triangle offense. Two, three times in a row with the same result: a three-pointer. How should the defense adjust without sacrificing even worse opportunities?
Here is the promised solution to how to make games interesting in perpetuity: let the plays the coaches are always calling out become public knowledge. Each team has a heavily guarded play book that lists all of their offensive and defensive schemes with all possible variations. These books are never shown to anyone outside the franchise, and when players are traded, their books are taken back from them. To counter the secrecy of the books, all teams send out advance scouts to determine from the games they play what the offensive and defensive sets are. In this way, every coach in the league ends up knowing every other coach's plays, and then they teach them to their players to show them how to break it down.
So who are these play books a secret from? Only the fans. It is only the fans that don't understand what is going on, or at least not as well as they could if that information were available. Sometimes, little nuggets float through, as nba.com analyzes the triangle offense or a simple pick-and-roll. Those little morsels are insulting.
Give us the playbooks!
Let the fans understand what's going on!
The coaches are not huddled over thinking, "Ooh, let's run a pick-and-roll this time!" "Great idea! I never would have thought of that!"
They have sets and possibilities and cuts and options far more subtle, complex, and beautiful. Let the fans pore over these playbooks to their heart's content. Let the people paying the salaries and bonuses of everyone in the game get a taste of the product they're paying for.
The entertainment value of watching basketball when you don't know the rules amounts to nothing more than waiting for a flashy dunk or three-pointer.
The entertainment value of watching basketball when you don't know all the plays and sets and options the teams are running amounts to nothing more than waiting for a pick-and-roll or a mismatch.
Free the playbooks. Or as the graffiti in libraries commonly says, "Free the bound periodicals!"