Thus, I was a little befuddled by this Malcolm Gladwell article. I've played a full court press, and my memory is that we got creamed. That's because my team was terrible. Indeed, as I recall, we abandoned the tactic after three or four games, because it wasn't improving our scores, and might have made them worse.
On the other hand, my team was really, really terrible. Perhaps a slightly better team would have been more successful. What do I know?
Via Kevin Drum, however, I learn that Chad Orzel has the same opinion, and he seems to know what he's talking about:
He goes on at some length about what a bold innovation up-tempo basketball is, and how it completely changes the game in favor of the scrappy underdog.This whole line of argument is somewhat undermined by the fact that there are several decades' worth of examples of people using pressure defense successfully in basketball. OK, he does make a nod to the history of the game by talking with Rick Pitino, but contrary to what you might take away from talking with Rick Pitino, pressure defense in basketball does not begin and end with Rick Pitino. Nolan Richardson, John Thompson, Billy Tubbs Paul Westhead, and others had success with pressing teams well before Pitino's success at Kentucky.
And, in fact, Pitino's 1996 Kentucky team is a terrible example of the point Gladwell is ostensibly making. Yes, they were a pressing team, but they were hardly scrappy underdogs-- nine of the players on the 1995-6 Wildcats went on to play in the NBA. This is not some scrappy group of nobodies who took it to the big boys with their unorthodox style of play. Quite the contrary, in fact-- that Kentucky team was loaded with top-flight basketball talent.
If you want to find real examples of people using the pressing style to overcome opponents with superior talent, you won't find many at the championship level. In fact, you're more likely to find stacked teams losing in spite of playing an up-tempo game than you are to find real teams of scrappy underdogs using the press to beat better competition. Arkansas and UNLV won titles in the early 90's with a pressure game, but more or less the same UNLV squad lost to Duke in a classic game in the '91 Final Four, and Billy Tubbs's Oklahoma team lost to a Kansas team that fits the "scrappy underdog" mold a lot better-- they had Danny Manning and not much else.
There's a reason for this: the press works, as long as the other team isn't ready for it. The idea of a full-court press is to force the opponent into a rushed and frenetic game and get them out of their routine. A team that's ready for it, though, and has skilled and disciplined players, won't get rattled by the press, and can pick the press apart for lots of easy baskets. You can use the full-court press to rattle a superior team that isn't expecting it, but if they know it's coming, there are a lot of ways that pressure defense can fall apart-- missed traps in the back court lead to two- or three-on-one breaks, over-aggressive defense leads to fouls, etc.. The teams that have won titles using pressure basketball have also had lots of talent, because you need something to fall back on if the press doesn't work.
Unfortunately, while you do get bloated incumbents who collapse when you use unconventional tactics, most companies, armies, and so forth are actually run by professionals. If they are good at their job, your unconventional tactics will make you lose worse than otherwise--think Netscape giving away its product for free in order to get widespread adoption, then finding out that Microsoft could do that for longer than a startup. The history of business is littered with groundbreaking innovators who were eventually shoved out of the markets they'd created by incumbents with complementary assets; most business school students will be familiar with the case of medical scanning, but the examples are legion.
Worse, unconventional tactics can trigger an adverse reaction in opponents. If they have the ability to punish you, you'll regret it. I suspect that most of Al-Qaeda were surprised and horrified when the US reacted to attacks on US soil differently from attacks on military installations abroad: not with surgical strikes on moderately important targets, but by invading Afghanistan. Similarly, if the players on the basketball court think you're going too far, as any basketball player can tell you, they have lots of ways to retaliate.
Wow, that Gladwell piece is almost self-parody.
Did he actually write a long article questioning why the best teams in the best leagues don't adopt the strategy that was proven to work in a 12-year old girls league by a parent-coach?
I think this proves that Gladwell is embarked on an Andy Kaufman-esque quest to see how self-evidently stupid he can make his writing before anyone dares to stop lauding his every word.
Steve Sailer has a good evisceration of Gladwell's argument "it's just Malcolm being Malcolm."
Who is Malcolm Gladwell?
Derek
The example from Kentucky that Gladwell should have used were the teams from 1991 to 1993, but even those teams were not untalented (that was a perception Pitino carefully nurtured in his opponents), but by 1996, Pitino had undeniably overwhelming talent.
The history of basketball shows that Davids beat Goliaths with ball control strategies. These worked better in the age before shot clocks in the college game- see NC State vs Houston 1983, Villanova vs Georgetown 1985, or Princeton vs anybody until the late 80s.
That's so wrong it's like an April Fools day joke in May.
First, the press is as old as basketball. Why does Gladwell think nobody would have thought of it like, oh, back in the days of the Auerbach Celtics? Coaches were supposed to be stupid back then?
Second, his idea directly contradicts logic, as demonstated both in today's quantitative sports analysis and classical strategy.
That is, the "uptempo" style of play creates more possessions per game, and more possessions per game favors the better team, not the underdog, other things being equal. The underdog wants fewer.
Consider a dice game where on each throw of the dice one player has a 67% chance of winning (numbers 1-4) and the other has a 33% chance of winning (numbers 5-6). The person who wins on the most throws wins the game. The "weaker" player will want fewer throws -- with one throw he has a good 33% chance of winning. But what will his chances of winning the most out of 100 throws be?
In basketball the stronger team averages more points per possession, so by the same logic it wants more possessions per game, other things equal.*
This is why the classic strategy for the weaker basketball team against the stronger has always been slow down -- to the point of the "four corner offense", in which the weaker team passes the ball endlessly until it gets one easy shot it can't miss. Thus maximizing its points per possession while reducing the total of them.
As to "Pitino and the press", hey, at little college programs like Kentucky and Louisville he's always had the BETTER players in 95% of the games he plays during the season. That's why the press has worked for him.
But what happened when he took that style of play to the NBA's Celtics when they didn't have superior players? He got clobbered and run right out of town. Right back to the college ranks where he was assured of having the better players once again. So much for the press empowering the scrappy underdog.
The most dramatic example of the super-press, super-aggressive game stragegy in the NBA was Paul Westhead bringing it from college to the Denver Nuggets. Yeah, he immediately zoomed the team's offense way up to 120 points per game! Wow. Unfortunately, they also gave up 131!! Including 107 in ONE HALF in a game against the Suns, which remains a record that may never be broken.
That's what happens when the team with the least talent ratchets up the number of possessions per game. Westhead went 44-120 before he got fired, one of the league's historic calamities.
So much for the press empowering the scrappy underdog! ;-)
(* Yes, yes, other things often are not equal, you have to adjust for the particular skills of the team you've got. Some good teams have a slower style of play. I grew up with the Red Holzman era championship Knicks, I know that.)
But what will his chances of winning the most out of 100 throws be?
Still 33%, assuming the throws are genuinely random. The number of throws does not change the probability, hence the danger of using them in the analogous way you propose. What the disadvantaged player really wants is a chance to throw the dice down the garbage disposal the very second he gets ahead after a brief streak of winning throws, because a large number of throws will trend back toward the probability ratio.
There is a misunderstanding in here somewhere (if I am reading this correctly). In any single throw, he has a 33% chance of winning, but if the dice were tossed 100 times, the odds of him having more wins out of those 100 than the other guy is much, much less than 33%. And in a million throws, his odds of having more wins is essentially zero, and if it isn't, then there is a high probability that the dice were loaded.
"But what will his chances of winning the most out of 100 throws be?"
Still 33%, assuming the throws are genuinely random.
Um, no. For "best of three throws" the odds change to 74%-26%.
I leave "best of 100" as an exercise.
Oh, okay, the probability is of winning for a total of n throws, not the odds of winning on each particular throw. I think I follow now.
@ aMouseforallSeasons,
That is incorrect. The law of large number indicates that the more times you "flip a coin" (or shoot a basket, or...), the more likely you will be near your average. That is why it is relatively easy to get "all heads" when you flip a coin only once, but almost impossible when you flip it 100 times.
Ok, the original answer (1/3) was wrong, but the right answer turned out to be fascinating.
Took me 10 minutes and half a page to demonstrate that for odd N (which are more interesting than even in the sense that they give higher probability of win for the underdog), the formula is:
p(N,1/3) = Σ{i=1..(N-1)/2}( C(N,i) ) / 3N
where C(n,i) are binomial coefficients. Next I tried to wiki and then google partial sums of binomial coefficients... not only there isn't an exact formula, there doesn't seem to be a good approximation that would not be hidden in an offline article somewhere. I didn't even find the exact asymptotic behavior. All I can say for sure is that it is going to be less than 1/3 :-)
Changing the subject to a sport I know a little more about...
The same could be argued of the NFL; offensive juggernauts that score quickly can be stymied by ball-control-offense underdogs. Problem being, if they expect ball-control, the stronger offense tends to be able to play ball-control just as well as the underdog.
"NFL offensive juggernauts that score quickly can be stymied by ball-control-offense underdogs"
In the NFL the most points per possession wins just like in basketball. Given equal quality defenses, the team with the best offense wants a lot of possessions and the team with the weaker offense fewer, just like in basketball.
So, yes, the weaker team may well go "ball control" to reduce the number of possessions and hope to get lucky -- if it gets a couple of lucky scores, or the stronger team has a couple of unlucky turnovers, the stronger team may not get enough possessions to make up for it.
But in the NFL defense is a major factor and has a rather different logic due to the human element and the nature of the game.
If teams have equal strength offenses but one team's advantage is a better defense, it may want fewer possessions. In football the defense must cover the entire field every play without knowing what's coming, while the offense attacks one point and knows what's happening in advance. So if the defense is on the field for a large number of plays and possessions, its players may get physically tired and its strategies may become exposed because the offense has more plays during which to look at them and figure out counters.
Thus a "defense first" team may play on offense to chew up the clock and reduce the number of possessions even if it has a very good offense, and even though this counters the general rule for the team with the best points per possession. (Lombardi's Packers maybe being the all-time example.) Things often aren't equal. (The superior defense might be viewed as at risk of giving up an increasing number of points per possession with a rising number of possessions.)
But in any event, none of this kind of strategizing is *new*. Red Auerbach's Celtics were collecting championships with the aggressive press-and-run long before Gladwell was born. When Pitino was in diapers. Of course, Red always had the better team.
It's natural for Malcolm Gladwell to believe that mediocrities can succeed by picking one simpleminded idea, and applying it to every problem in sight.
[Kent_Brockman]"And I'm going to have to say...Ouch! for Malcom Gladwell."[/Kent_Brockman]
Gladwell has a point, he just overapplies it. The point is that if you are outmanned, the element of suprise can be a force multiplier. Little girls basketball teams do not usually run the press because little league basketball is about teaching fundamentals and making practices fun so they don't quit. Thus, the other teams were suprised. In higher levels where there is scouting and the players are serious, running the press will not suprise anyone and has no great advantage.
The problem with applying this lesson is that the element of suprise is very hard to use against forces that are smart and prepared.
The main point of Gladwell's article is that underdogs give themselves a fighting chance with unorthodox tactics. They may not be able to win the war or championship over the long haul, but they could win the battle or game. Certainly it makes more sense for a coach that enters a contest with an inferior team to try to come up with an innovative approach, rather than to hope that his team plays the perfect game. Hope is not a strategy.
As they say in boxing, styles make fights, and an unorthodox style can stymie a far superior fighter. However, most coaches/business executives are afraid to do something unorthodox, so they end up losing the game/account by playing it safe and getting pummeled.
A secondary point is that Rick Pitino is probably best suited to coaching 12-year-old girls.
If I may ask a meta-question, why are you even wasting your time commenting on an article by Malcolm Gladwell? Isn't Gladwell's shtick sort of old by now? He gets paid lots of money for writing happy talk about how no one has any inherent advantages (e.g., height, or athleticism, in the case of basketball) and greatness is all due to elbow grease.
Is there no way to get you, as a prominent econoblogger, to read and respond to the writings of an expert financial practitioner and Stanford-educated Ph.D. economist such as John Hussman?
It must be true, because the model works in computer simulations.
Those simulations being any version of EA Sports' NBA Live: full court press, control a player with good stealing attributes and just have him stand between the inbound passer/passee while frantically pressing the steal button, which is the same button for a dunk/lay-up. You'll win by at least 100 points, guaranteed.
Which is why, if you're an innovator, it's important to have exit strategy plans which including options for selling out to a larger company with deeper pockets and market access.
Not an easy thing in a recession.
Which is why, if you're an innovator, it's important to have exit strategy plans which including options for selling out to a larger company with deeper pockets and market access.
Heck, if you don't have that, you'd better have a hell of a house to borrow against, because you'll NEVER get angel or venture funding in the age of SarbOx, and patent lawyers don't work for free.
According to the Center for Venture Research, '08 was down in $s but not deals (angel investing); mergers & acquisitions were 70% of the exits, and returns varied but averaged 22%; 26% of the remaining exits were bankruptcies. Reports here
There is also the example of the decidely low-tempo, fundamentally sound offense favored by Princeton University which has had substantial success against far more athletic teams in the NCAA tournament.
Here is the best list I could find of Princeton guys who played in the NBA (I cannot attest to its veracity). 10 guys, or 1 more than Kentucky had on just their 1995-1996 team. Most of these guys were from a prior much less competitive era, so clearly a group of athletic underdogs.
Bradley, Bill 1967-1976 HOF
Goodrich, Steve 2000-2001
Hill, Armond 1976-1983
Hummer, John 1970-1975
Kearns, Michael 1954-1954
Manakas, Ted 1973-1973
Palmer, John 1946-1948
Petrie, Geoff 1970-1975
Taylor, Brian 1972-1981
Vanbredakolff, Butch 1946-1949
Brain Taylor was a very good player. Guard on the old Nets teams with Dr. J, Larry Kenon, John Wiliamson and the Whopper (Billy Paultz) at center. They were a great fun team. ABA champs a couple of times iirc. Not NBA championship quality but frankly the ABA ws much more fun to watch. But definitely not a"low-tempo fundamentally sound" league, quite the opposite, which made for much of the fun.
Geoff Petrie was pretty good too. Vanbredakolff had a very successful coaching career.
I also think one of the most important elements in having a successful press is a very talented, long, athletic shot-blocker to cover up for blown assignments. These type of premium athletes are highly unlikely to be on a under-talented team.
I always thought the point of a full court press was to wear down a technically skilled but less athletic team. Of course, the last team I was on (in 7th grade) lost every single game in the season.
I suspect that most of Al-Qaeda were surprised and horrified when the US reacted to attacks on US soil differently from attacks on military installations abroad: not with surgical strikes on moderately important targets, but by invading Afghanistan.
And even more surprised when, in the face of guerrila IED attacks and massive suicide bombings in Iraq, we didn't retreat as in Vietnam and Somalia but instead successfully turned the Sunni tribes against AQ by moving from a force-protection stance to securing the safety of the local civilians as a first priority.
Because we removed the Sunnis from power, AQ assumed they would always hate us more than they could hate AQ, but a few years of contact with both sides quickly led Sunnis to find us the much lesser of evils. Fallujah and Ramadi went from being places where our guys couldn't ever leave their bases without suffering multiple attacks to a couple of the safest towns in the country.
Someone at al Qaeda must have predicted the response. Otherwise why assassinate Ahmed Shah Massoud only days before 9/11? I suppose it's possible that that was a coincidence but it doesn't smell right.
The Gladwell article contained interesting anecdotes but all the inferences he drew struck me as really stupid. For example, his main point is that effort can trump ability. But all the examples he uses involve the defeat of purely physical ability, not merely by effort but by mental ability - actually engineering skills. The most obvious example is David beating Goliath. David's physical effort appears relatively light. What David did was engineer a solution. Similarly, the basketball example is not success based solely on effort but also on insights about how the rules of the game lead to certain patterns that are vulnerable to a counterstrategy. The real point I think is that engineering skills can amplify effort to overcome raw physical attributes in competition.
Megan,
The examples in your analysis are completely erroneous. First:
If they are good at their job, your unconventional tactics will make you lose worse than otherwise--think Netscape giving away its product for free in order to get widespread adoption, then finding out that Microsoft could do that for longer than a startup.
Contrary to your assertion, it was Microsoft's groundbreaking idea to give away its browser for free (forcing Netscape to follow suit), for the exact (anticompetitive?) reason you state.
Second:
Worse, unconventional tactics can trigger an adverse reaction in opponents. If they have the ability to punish you, you'll regret it. I suspect that most of Al-Qaeda were surprised and horrified when the US reacted to attacks on US soil differently from attacks on military installations abroad: not with surgical strikes on moderately important targets, but by invading Afghanistan. Similarly, if the players on the basketball court think you're going too far, as any basketball player can tell you, they have lots of ways to retaliate.
I don't have the slightest idea what you mean by an adverse reaction from an opposing basketball team (a melee incited by a press? what?) but your read of Al-Qaeda's intentions with the 9-11 attack are way off. OBL and his affiliates have explicitly stated that they desired to engage the USA in an overreaching response that would bleed its resources, just as the USSR bled in its own Afghan war.
My God, Gladwell REALLY REALLY doesn't know what he's talking about!
I wasn't able to link to his article through here before, and just read it.
That Fordham team he cites as evidence for his thesis -- those "scrappy street kids" who by hustle and pressing beat a U Mass team at U Mass -- hey, I knew that team personally!
Those "scrappy street kids" included two future NBA players (including a first-round NBA draft pick), went 26-3 on the season, was ranked in the top 10 in the nation, and played the nation's #1 team, Marquette, dead even into overtime before a sold out Madison Square Garden.
How the hell does Gladwell come off selling them as an underdog that won only by dint of surprising hustle?
You'd think that before describing one team as an "underdog" to the other, he might look at which one was nationally ranked and which wasn't.
Remind me to never buy any of his books.
And remind him never to believe retired coaches with self-indulgent memories: "Aw, they were just a bunch of street kids I inherited, but with my brilliant coaching..."
Yep, if you've got a skill imbalance you generally want to slow things down and give the other team fewer chances to demonstrate their superiority. In the era before the shot clock there were some cases like a 1981 game between Stanford and Oregon State with a final score of 18-16, in a Division 1 PAC-10 league game. Stanford was very much the underdog so they simply held onto the ball. The second half was the high scoring one in that game as I recall. I think the first half score was in the single digits.
A good passing team can pick apart a press and get a lot of layups.
For the record as well, the Gladwell's analogy to soccer is equally spurious.
No team in the world plays the equivalent of a "press" in soccer. It would be insane given that at the highest levels field players already run 5 to 10 miles a game playing a loose defense.
When a soccer team applies "high pressure" what it means is that the forwards play defense and chase aggressively around midfield. The tried and proven method for underdogs to compete is to drop 11 players behind the ball and play for the counterattack. It's worked wonders for the US over the last few years and made Greece European Champions in 2004.
Now it's also possible to acknowledge that Gladwell has a point: David's win by not playing Goliath's game, but forcing Goliath out of their normal style of play. Where he errs is not recognizing that the press might be one way of doing that against teams with poor ball-handling but the Princeton offense is the way to do it against athletic teams who play aggressive defense, etc. etc.