clock menu more-arrow no yes mobile

Filed under:

It's Better to Be Lucky *and* Good - Rain Delay Bonus Content!

I recently took a look at a milestone in the season - the "one-third" mark after game 54. The dividing line was the game prior to the 10 - 2 loss to the Nationals on May 31.  Through this point, one third of the season is over, and it may be a good time to assess what we are all seeing.

The Phillies have the best record in Major League Baseball.  Is this attributable to skill?Are the Phillies objectively an elite team in 2011 or are they just lucky so far?  The answer right now, it appears, is "both."  How can anyone tell if it is luck or skill?  This post will help try to explain how some people try to separate team "luck" from team "skill" and then apply that analysis to the Phillies YTD through May 30, 2011.

Most readers of this site are familiar with the fickle BABIP Fairy, regression, and ERA vs. xFIP.  We know that some outcomes are lucky and unsustainable while others are absolutely true.  There are three widely-recognized "types" of wins that help us understand, albeit imperfectly, in the context of team performance, what is real and what is smoke and mirrors.

"First order wins" are the regular old wins you see in the newspaper.  "Second order wins" are those adjusted by some mathematical sleight of hand to try to sort out "lucky" versus "good."  "Third order wins" are second order wins adjusted for the strength of schedule that a given team has played.  The idea is that third order wins, in particular, should give a more accurate picture of a team's performance than pure wins/losses.  Three wins against the Astros don't say as much about a team as three wins against the Red Sox, for instance. The ideas behind second order and third order wins are explained really well here.

Adjusted standings are available from Baseball Prospectus here.  A review tells us that the Phillies, while very good this year, have been among the luckiest teams in baseball, if second order wins are based on good statistics that measure all things in baseball accurately. You can see that only the Orioles and Indians have outperformed their first order wins by as much as the Phillies so far (again, through May 30th -- the report at BP linked in this story will change over time).  The third order wins (adjusting the records for strength of schedule) show that the Phillies collectively have "outperformed their peripherals" more than any other team in baseball through May 30.  

This does not mean that the Phillies are Kyle Kendrick waiting to crash back to replacement level (even good players can outperform their peripherals -- this does not mean that they are bad, just not as good as they appear at first blush).  It does mean that talk of a 100+ win season is premature.  And to be fair, the Phillies strength of schedule will be weaker than, say, the Nats, because the Nats don't get to play the Nats -- they have to play the Phillies instead.  The flip side of that helps out the division leading teams, whose strength of schedule will inevitably be lower than trailing teams in the same division.  This "hurts" the Phillies' third order wins a little, but any distortion caused is less than that which would be caused by not adjusting a team's wins for strength of schedule.

The second thing that jumped out from the adjusted standings for me when I wrote this was how unlucky the Braves have been.  They are the second "unluckiest" team in baseball with a -3.4 win differential (again, through May 30).  The Braves, if all luck was perfectly evenly allocated in a world with a completely neutral strength of schedule, would be in first place.  The adjusted wins bear out my personal theory that it is the Braves, not the Marlins, who present the biggest the threat to the Phillies this year, even though the Braves have spotted the Phillies a moderate lead during the first third of the year.

Keep in mind that there is no guarantee that the Braves will have "good" luck from here on out since they have already had a disproportionate share of "bad" luck.  Strength of schedule may even out some, but third order wins aren't really giving rise to the "problem" here in terms of "distorted" standings.  In addition, the Phillies are equally as likely to have more "good luck" as they are to have an offsetting amount of "bad luck."  Statistics will not operate over the remaining 108 games as a mathematical karmic wheel of justice.  Just as players are not "due" over short periods, neither are teams.  

It is really hard for a team to dig itself out of a hole.  One of the reasons is that there are essentially four plausible scenarios for luck the rest of the way.  It's pretty obvious, after looking at them, why it is unlikely that the "luck" factor will equalize over the balance of the year. That and the fact that what has happened so far with respect to luck has absolutely no bearing on what comes afterward.  Again, "luck" is not a karmic wheel of justice.  It's just luck.



  1. Braves and Phillies both are extra lucky (draw)
  2. Braves and Phillies are both extra unlucky (draw)
  3. Braves are less lucky than the Phillies (+Phillies)
  4. Phillies are less lucky than the Braves (-Phillies)
Three out of the four scenarios favor the team with the advantage in the status quo, which is the Phillies.  The Braves need a changed state, and only one outcome will allow that.  This ignores non-luck factors that can close the gap, such as objectively better performance, getting better players by trade, or injuries, etc.  The above is just related to the luck aspect.


Thoughts about second and third order wins aside, the Braves aren't in first place, and the Phillies are.  MLB recognizes only first order wins.  That the Phillies have performed as well as they have with huge chunks of games played by WAR criminals like Valdez and Schneider speaks volumes about how good their pitching has been.  While that has been discussed thoroughly recently, I have another spin on it to offer in a couple of days.

For the purposes of today, console yourself with the fact that the Phillies have the best record in baseball (so far).  It is largely deserved, though a small portion has been the result of luck.  And that record was achieved with Chase Utley, Dom Brown, Carlos Ruiz, Brad Lidge, Jose Contreras, Joe Blanton, and Roy Oswalt all missing significant time.  With some of those players back, perhaps that 100 win talk may be justifiable.  

While I would be pleased to see a 100+ win season by the Phillies, I am still more inclined to believe that the Phillies will end up in the low to mid nineties of wins, as I predicted earlier this year mainly as a result of my discounts for injuries and age.  I think the Braves will start to catch more breaks and will be right there at the end because, dang it, they are really good, too.  But what fun would a baseball season be without cardiac events in late September?