ACE ON BASE

Moneyball (The Art of Winning an Unfair Game) is a book about baseball by American sport's writer Michael Lewis.

The thrust of the book is that The Oakland Athletics, a team struggling against the big money advantages enjoyed by the likes of the NY Yankees, Boston Red Socks, Atlanta Braves have been able to stay highly competetive. They could not afford to buy, say, Pedro Martinez or Randy Johnson for gazillions of dollars and, instead, had to look extremely closely at the statistics of lesser credentialled players such as Nick Swisher and Scott Hatteberg in an attempt to keep up with the big boys.

Up to now it has proved a reasonably successful strategy.

The architect of this ploy that's taken The As from a struggler in the late nineties to a potential contender in the naughties is former prodigy Billy Beane. A precocious talent as a youngster who apparently didn't have the right attitude to cut it in the big leagues as a player, but who is seen as something of a wunderkind as an administrator.

The main statistic according to Beane and his off-siders, is Percentage On Base. A stat which refers to the ability of a batter to get himself into play regardless of whether he hits the ball, or not. Key to this is the batter's self discipline in being willing to wait for the right one to hit, and just as willing to accept a walk. Hatteberg is a gun here. It's a batter's equivalent of playing the percentages.

Also covered in the book is the notion that while raw talent is undoubtedly a strong pointer to whether a player will make it in the big leagues, there is much to be gained from in-depth analysis of the less obviously talented draftees.

This, of course, may -- that is may -- have ramifications in Australian Rules. I stress may, because at The Athletics, a batter cannot continue to swing at bad pitches because Beane will sack him as quick as you can say "three strikes and you're gawn!" That can't happen here. Nevertheless, the book is one that hasn't escaped the attention of quite a few pundits in Aussie Rules. The latest is Gerard Healy in today's Herald Sun ...

Statistics add up

MONEYBALL, written by sports journalist Michael Lewis, is a must-read for anyone working in the sports industry.

It's about how a one-time failed star recruit Billy Beane became a manager of Major League baseball side the Oakland As and turned a poor, down-and-out franchise into a raging success.

At the time most put it down to luck, but those on the inside knew it was all about mathematics and a new way of thinking about the sport.

Beane didn't recruit big names – they couldn't be because Billy didn't have any money – but he picked up players that statistically fitted his analysis of what won games.

I read this book 18 months ago and have read it again since, and following an All-Australian team selection meeting last week I began to think of Moneyball again.

Healy goes on to make a case, using statistics, for compiling his list for All Australian honours.

I, too, read the book last year and found it to be an enjoyable, easy read. And while I agree with it's premise (different stats matter) I find I'm getting increasingly tired of the experts here referring to it without caveat.

Sure, stats do matter. But the way of looking at stats in Moneyball cannot be done so in isolation, and then directly transferred here. It must be seen as extra information to what they already know about a much smaller talent pool.

There are two main flaws associated with using the strategies of Moneyball in Australian Rules. One, hitting analysis is not comparable to the analysis needed in a contact sport. All it does is engender new streams of thought. And two, I've never yet heard one pundit here state the bleeding obvious. Namely that The As, while they've managed to win more than 90 games (55%) in each of the past five seasons, they haven't won The World Series in that time. Nor have they even made the play-offs every year. On the other hand, one of the players The As traded away is Johnny Damon. Damon has become an integral and popular member of the Boston Red Sox (a much derided team in the book) who, in fact, after more than eighty years HAVE managed to win a World Series.

And rest assured, if Beane's strategies prove successful, they won't be his alone for long. The big clubs will adopt them and everyone will be back to square one.

Comments

[Andrew]

This book seems to be creating quite a splash. Over here in the US it is popular to quote it. I think one of the other interesting things is the way that Beane doesn't like to draft good high school players and would rather the kids develop in college or at another club before he looks at them.

You're right that the As haven't won any pennants, but it's good that he sets them to be a near-abouts contender every year, bacause every now and then there is a chance that they can fluke a series.

Also keeping the team solid on the field while making sure the front office stays in the blue means that a strong base is built for the club to develop on.

Nice to see that an Aussie having a decent grasp of ball though. I have no ideas at all about your Aussie Rules.

Good stuff.

[Tony.T]

Yeah, that's fair enough, Andrew. My main concern though, is that while The As relentlessly "play the percentages" they miss out on that special "something" that allows a side to win a World Series.

It seems the likes of the D-Backs, Angels and Marlins are the ones fluking the World Series, not the As.

[James Dudek]

A closer read of Moneyball will show that one of the disciples of Beane's methodology was a young Theo Epstein, who would go on to become the general manager of the Boston Red Sox the year they won the World Series. At the time Moneyball was written Epstein was a scout for the Red Sox and it was his influence that stopped Kevin Youkalis being traded to the A's, much to Beane's annoyance.

A closer reading will also show that John Henry who became owner of the Red Sox the year before they won the World Series is a disciple of this methodology. Interestingly John Henry made his fortune using a similiar statistical approach in futures trading.

It was the combination of the sabremetrics approach and having deep pockets that allowed the Red Sox to REVERSE THE CURSE!!

[Russell Allen]

Baseball is like an eggtimer for beer drinking, no more, no less. When H. Simpson went to the baseball sober he wanted to top himself after the second inning. Fact!

What's that got to do with stats I have no idea but I wager that 9/10 of the population of the world reckon that if the US allowed other countries into the World Series they'd lose - like all the time ya'know. I saw that movie Mr Baseball, with Tom Selleck - those Japanese can play...and the Puerto Ricans aren't too shabby either.

[Tony.T]

And not forgetting, James, that Paul De Podesta was offered the top job at the Blue Jays but said no, so they took J.P.Ricciardi from The As instead.

And Beane, himself, was offered the top job at The Red Sox and had all but signed but knocked it back at the last moment.

[vague]

The only thing that bugs me about this, really, is that the commentators always make the "it's really your on base percentage that counts" analysis like it was a huge effing revelation. It's common sense--of course you want hitters who get on base.

As for the As, they should be doing a lot better than they are, but I find it hard to care. If I have to watch a western division team, I'd much prefer it to be the Mariners. At least as long as it isn't the Giants; I just loathe them.

[Tony.T]

Very true, V. It's why I mentioned the "stats in Moneyball cannot be done so in isolation". Sure it's good to get on base, but it's just another important part of an overall game strategy.

If all you are concerned with is waiting for the right one to hit, you'll probably always struggle against really good pitching. And when are you going to see the best pitching? When it matters. That's when. The playoffs.

And I like to see the Giants get beaten too. Dunno why. It's not as if they're a regular contender like the Yankees. But I just don't like them.

[Tony.T]

Oh. And by the way. I write as someone who only gets to see one game a week, so feel free to suggest I could very well be talking out of my ar ... err ... hat.

[vague]

They're impossible to like. I blame Barry Bonds and his stupid earring.

[Tony.T]

I'm continually puzzled as to why men with stupid earings aren't in jail.

Naturally I barracked to Anaheim against the Giants in 2002. And that's in spite of that stupid monkey!

[Alistair]

A friend of mine pressed his copy of Moneyball into my hands recently. "You must read this," he said.

Interesting that you think Aussie Rules might benefit from similar analysis. Would cricket not be more useful a transfer?

[Tony.T]

Perhaps you misunderstand me, Alistair.While I don't suggest exactly the opposite to what you write, I'm certainly sceptical about Moneyball's effect on Aussie Rules.

The book is repeatedly brandished by Aussie Rules pundits here as a marvellous pointer to a new horizon in statistical analysis of the game. With many saying exactly what your mate said, "You must read this!"

My point is that the comparison is not everything it's made out to be. That's why I wrote "hitting analysis is not comparable to the analysis needed in a contact sport." Sure, it's food for thought, but only as extra information to a game already bulging with statistics. Anything gleaned from MB would merely be "extra info". NOT a bold new era in the way Australian Rules is statistically quantified.

Cricket would, indeed, be a better comparison.

A specific example would be when New Zealand toured here in 2001/02 and their bastmen virtually refused to play Glen McGrath unless he put it right on the stumps. Leaving most everything go is cricket's equivalent of baseball's "playing the percentages" by not swinging at bad pitches. Matthew Sinclair was twice bowled leaving balls that were too close to off stump.

This tactic made New Zealand very competetive. They drew the first two tests and almost won the third in Perth. However, if the whether hadn't intervened in the first two tests, robbing Australia of almost certain victory, NZ would have been comfortably two/nil down going to Perth. Bear in mind also, for all that NZ played well in Perth, they needed a huge slice of luck to stop Gilchrist and Steve Waugh pinching a superb come-from-behind victory.

[Alistair]

Sorry, Tony, didn't express myself very well. I was surprised that Aussie rules was being held up by *anyone* as ripe for Moneyballing. I don't know anything about Aussie Rules, but I would guess that it is harder to analyse than baseball (essentially a series of one-to-one encounters). But you say that it is stat-ridden, so I’m obviously talking bollocks. I have a hunch that the market in football (soccer) here may be distorted - certain skills are overvalued, people who look good scoring goals, or people who score the occasional great goal, are valued more than efficient, ugly players - but it's so hard to measure in contact, flowing team games.

The book hasn't had anything like the same effect here. I heard of it when a sports writer picked it as one of his books of the year; otherwise it has yet to appear in the public consciousness.

[Marty]

"My main concern though, is that while The As relentlessly "play the percentages" they miss out on that special "something" that allows a side to win a World Series.

It seems the likes of the D-Backs, Angels and Marlins are the ones
fluking the World Series, not the As."

From a GM PoV, the postseason is exactly how Billy Beane describes it, "a crapshoot". When the Angels won the World Series back in 2002, they finished second to the A's in the AL West, 4 games back.

Given the small sample size (number of games) in a World Series, I feel far happier saying the A's were a better team that year, as they proved over the 162 game season.

The Postseason is won and lost by the form of individual players - there is little a GM can do, other than the same thing he does during the regular season: arm his team with the best players he can buy.

Baseball is a far closer sport than many others - a great team wins 6 out of 10 games, a terrible one wins 4 out of 10. Compare that to the Premiership, where Champions Chelsea only lost 1 game, and previous Champions Arsenal didn't lose any over the 38 game season.

This means that in a series of a few games, it's far more likely a terrible baseball team can beat a great one, than a great football team losing to some filth.

So in summary, the World Series Champions may not be the best team in baseball that year. Shock!

[Clem Snide]

In a true team sport, stats for individual players are mostly bollocks, since they fail to account for the synergistic effects crucial to success. I think they are mostly used to give commentators something to talk about. Cricket and baseball are only illusory team sports, as they are actually a series of solo performances which start from a set play. The solo aspect of these games removes synergistic effects, while the set play aspect enables comparison of like with like, allowing for statistically significant samples with high signal/noise. The most meaningful stats in AFL/ARL/Union are those for set plays, e.g. kicking distances & goal-kicking percentages. Even solo stats like yardage gained, number of possessions & number of tackles are hopelessly contaminated by field position and team feeding ability. That's why I think that the gang-bang and binge-drinking bonding sessions which have got Thugby players into so much trouble are probably far more important to success than stats. Admittedly AFL is more based around set plays than other football codes, but even so much of the important stuff is scrappy and team-based.

Baseball is perhaps the optimal statistical sport, given the limited variety of batting, pitching, running and fielding options available to the players compared to cricket, and the lack of necessity for all-rounders. For cricket, raw stats aren't enough. For instance with the Australian team, the assistant coaches punch every delivery into a computer with details of where the ball was pitched, how fast it travelled, shot played, where the ball went after it was hit etc. Thus they can rigorously evaluate the near-misses and skills deficits of each player. If someone could develop a method for rigorously measuring and analysing interactive aspects of team sports, they would make a fortune.

[James Dudek]

NE Patriots coach is also a big Moneyball man. For example, statistically it's better to go for the first down when your on fourth down. Conventional wisdom around the NFL is that you always play it conservative. Because Belichick is willing to look like a fool, he'll go for it and take the heat if it doesn't come off.

This has to be a direct correlation with Rugby League right?

I'm not sure about the AFL connection - too much fluidity.

[James Dudek]

Slightly off topic. I think the World Series is the perfect name for Baseball's trophy. Players from all over the world are coming together to compete at the very highest level.

In fact I would say that the World Series is a more appropriate name for the competition than the World Cup is for Soccer.

So what if one country can put together a team that beats another country. Isn't it supposed to be about who has the best combination of players regardless of nationality? You're telling me that Real Madrid couldn't beat the Brazilian national team?

It's one of my pet peeves that people think that by calling it the world series the Yanks are dismissive of any other nation. Japanese, Colombian, Mexican, Dominican Republicans, Aussies, Canadian, Venezualans, Chileans, etc, etc, etc players all populate the MLB.

[vague]

Word, James Dudek. Word.

Also, Tony, I like that you call it "a game already bulging with statistics." It never fails to amaze me what kind of nonsense the commentators can pull out of their, um, hats while a batter is fouling off his fortieth pitch. "Full count...and Ramirez fouls again. You know, Carl, in games played on a Tuesday where the Sox were down 4-2 in the seventh inning, Ramirez has previously batted .567 when facing right-handed pitchers when the sun was out...Still full count." It's hilarious.

As far as this playing the percentages, I had another thought. I think the reason I prefer to watch say, the Mariners rather than the As is because it's just more fun to see players like Ichiro Suzuki, who gets a hit practically every time at bat, steals loads of bases, and is one of the best outfielders I have ever seen play. Just watch him reach backwards over the wall to steal a home run, and look at the batter's effing shocked expression--it's gold. Much better than watching dudes strike out all the time waiting for the picher to throw right to their sweet spot.

[Bruce]

So what has happened in the last 18 months. Stats generation has been growing at an amazing rate for both footy and cricket. Heat maps and pitch maps are making their way into the coverage and if they are being used for color and analysis by the media then they are almost certainly entrenched in the coach's boxes.

Moneyball has been good for me at work as something I can invoke when telling American bosses that the things they are measuring are not necessarily indicators for success or failure.