From ABC News: ‘On a night when principal owner John Henry was moved to say, “We’re not just a bunch of stat geeks,” the Red Sox …’ What exactly was he trying to say there?
Moneyball
The Red Sox are one of the leading baseball organizations embracing advanced statistical analysis of baseball players. They go well beyond looking at things like BA, HR and RBI totals to get at the core of what makes a “good” baseball player and what that player should do to contribute to a winning team. Go read up on things like VORP, Win Shares, Adjusted OPS+. It’s great stuff, although once you read it you will be forever unable to listen to Joe Morgan prattle on about how important “Wins” are to a pitcher.
It is great stuff. I am just perplexed at how that form of analysis leads to signing coco crisp, jd drew, julio lugo, and eric gagne for upwards of $33 million. throw matt clement in there and you have 90% of the rockies payroll. Do their win shares stats improve even if they are on the bench?
>I am just perplexed at how that form of analysis leads to signing coco crisp, jd drew, julio lugo, and eric gagne for upwards of $33 million. In my opinion, because BOS can afford to do so. If expensive players flop, big deal, make them back up or eat most of money and trade away. Teams like NYY and BOS have big margin for error. >throw matt clement in there and you have 90% of the rockies payroll. Another aspect is that they paid $51 mil just to negotiate with Matsuzaka. Rockies entire team pay roll this year is $54 mil. Not that I think Matsuzaka flopped/will flopp though…
What’s your point? It’s going to be tough to say the RS strategy isn’t good the day after they win the WS for the second time in 4 years. Let me guess, I bet you REALLY like David Eckstein?
my point is simple the success of the RS is as much a function of resources as it is their ‘superior’ analysis of talent. simply put, if you have the resources you can sign talent. you can value that talent however you want. going back to 1996, the yankees or rs have competed in 8 of the 11 series and won 6 of them.
Who would you have rather had on your team in 2005: Johan Santana or Cy Young award winner Bartolo Colon?
04 cy young winner santana or 05 winner colon… who would I rather have had in 05? They both had exceptional years. I would argue santana had better #'s but they gave the cy young award to colon because he won 5 more games. I would have been just as happy with either result. The question should have been: if it was oct. 2005, who would I have wanted on my team in 06. Based on 2005 numbers I would value santana higher for 2006. In 05 both made 33 starts and colon had 5 more wins but santana gave up fewer hits, fewer runs had a significantly better walk to so ratio, a lower whip and a lower era. So that would be my 2 minute decision. … and then Santana won the AL Cy in 2006. Who would you rather have on your team in '08: crisp or ellsbury? cora or pedroia? rivera or paplebon? Well if you are the RS you might want all 6… and you can pay all 6.
boo red sox
But Colon’s extra wins were in large part a function of what happened when he was off the field, whereas everything you mentioned about Santana was directly due to what happened on the field. I don’t see why people in this day an age (and especially participants in this industry) can even care about “wins”. It’s like judging a company’s performance based solely upon a net income number inclusive of all items. I don’t mean to just harp on wins here, batting average and RBIs suffer the same problems, and don’t get me started on judging fielders by “fielding percentage”, perhaps the most useless stat ever invented.
I don’t disagree. In fact that is exactly what I said. Colon won the cy (because they like to award that to 20+ game winners) but at that time based on 05 I would have taken Santana over him due to his numbers… Anyway that wasn’t the point here. The point was I don’t think you can attribute the sox recent perf to personnel gains due to superb “advanced statistical analysis”. First, I point to the perf of a number of the players they have signed recently in response to this analysis. crisp, lugo, drew, gagne. If stats were the deciding factor in those deals they might want to modify that model. Fact is, they have the resources to make deals. They can eat four bad moves with the hope of finding one good one (matsuzaka - maybe). 90% of the other teams can’t make those five moves. They have to pick one, maybe two, and that raises the risks of signing one of the busts. Second, does this qualify as advanced statistical analysis? Are we talking about simply using a different statistic to measure performance? Anyway, this started with John Henry’s stat geek comment and as you probably know from his wiki : "John W. Henry & Company, Inc. was established in 1981…after poor investment performance in 2005 through mid-2007, assets under management fell markedly from a high of $3.8 billion in 2005 to approximately $500 million in mid-2007. The firm employs methods of investing based on systematic, mechanical trend-following, meaning that the company does not allow for human interaction based on so-called fundamentals or human emotions.[4] On May 29, 2007, Bloomberg News reported that Henry’s assets under management had fallen by 80% after losing 33% over the prior two years. " …so I find it interesting. I wonder if the “systematic, mechanical trend-following” and overlooking of “so-called fundamentals” is reflected in the philosophy we are discussing here?!?
If a team like the Pirates can use the model and win the series then we would have to say they are on to something.
Well, the single biggest proponent of sabrmetric analysis is the Oakland A’s, who have been consistently competitive in a division filled with big spenders despite having a bottom quartile payroll suggests there’s something there. The Sox moves you described were perhaps not solely governed by stats. They don’t operate in a vacuum; they need to field a team from the players available. They can’t sign just whoever they want , and any given decision will factor in not only the “stats” of the players involved, but the contracts of the players involved, team needs, team payroll, to name just a few.
HoldSideAnalyst Wrote: ------------------------------------------------------- > Well, the single biggest proponent of sabrmetric > analysis is the Oakland A’s, who have been > consistently competitive in a division filled with > big spenders despite having a bottom quartile > payroll suggests there’s something there. > True. But the A’s spent north of 80 million this year which is middle of the pack not bottom quartile. They finished 10 games under 0.500 and missed the playoffs. I concede they are consistently competitive, but they compete head-to-head with Seattle, Texas, and the Angels to advance to the playoffs out of the AL West. Not exactly perennial powerhouses. And competitive is nice but where are the rings? 1989? I would argue that the three year A's explosion ending in 1990 was to sabermetrics what survivor was to reality tv. Without it, the practice would still be around, but without the fanfare, praise and copycats. \> The Sox moves you described were perhaps not \> solely governed by stats. They don't operate in \> a vacuum; they need to field a team from the \> players available. They can't sign just whoever \> they want , and any given decision will factor in \> not only the "stats" of the players involved, but \> the contracts of the players involved, team needs, \> team payroll, to name just a few. Agreed. As I have expressed I feel it is fair to say that the RS do not rely on advanced statistical analysis to make personnel decisions. I know that believing they do makes theo more attractive to rsn and makes them feel smarter than their evil rival but the reality is that they follow the same approach. I would sum up the RS and Yankees approach as 1. define the need, and 2. throw %#@ against the wall and see what sticks. I guess it is easier to explain a terrible decision by stating that the error term is unpredictable. You can always blame the model no matter if the measure is VORP, or RBI’s. That and it sounds complex to the fans and demands a salary premium.