2013 Playoffs, First Round - Part I
Joe Peta
2013 Playoffs
Why No One Should Heed my Post-Season Analysis: After a more than two-year immersion in all things Major League Baseball, once May rolled around this year, my attention got pulled away from baseball in general, and my spreadsheets in particular. I can read the numbers and enter the data just as I have any other year, but the truth is I simply don’t have as strong a grasp on the subtle trends of each team. Last year, for instance, I had an excellent read on the vastly underrated SF Giants offense due to second-half trends largely masked in the overall numbers by the team’s first-half production. That served me very well in handicapping the playoffs. That simply won’t be the case this year. Let’s put it another way: In everything I’ve written over the last two years – from full-season outlooks to single game previews – when I’ve expressed strong conviction I’ve have backed that up, without reservation, at sportsbooks in Las Vegas and online. There is no way I will demonstrate that same level of conviction this post-season.
Why it May Not Be a Bad Idea to at Least Consider my Post-Season Analysis: A data-centric sports blog named Stats in the Wild took a look at seven different pre-season team-by-team outlooks made by different projection systems and computed the forecasting error of each outlook. The very blog you are reading, frankly, dominated the contest with the smallest forecasting error under two different methods and the highest correlation between projected team wins and actual team wins. The other forecasters are the engines behind projection work at FanGraphs, ESPN, Baseball Prospectus and others. http://statsinthewild.com/2013/09/30/stat-pundit-rankings-mlb-win-over-unders/
National League Divisional Series:
Pittsburgh vs. St. Louis
A Look Back at my Pirates 2013 Projection vs. Actual Results:
W-L Record: 79-83 Actual: 94-68
Runs Scored/Allowed: 659 677 634 577
A Look Back at my Cardinals 2013 Projection vs. Actual Results:
W-L Record: 90-72 Actual: 97-65
Runs Scored/Allowed: 770 686 783 596
Although I got a handful of teams very wrong, Pittsburgh is the only one I wish I could take back. It would waste too much space to detail the research but suffice it to say when the Pirates had the best record in baseball after they’d played half its season on July 1, I was kicking myself. The Pirates were 51-30 at that point on the strength of the best defense in baseball. That was a trend I saw coming, dating back to 2011. The Pirates had the worst defense in baseball in 2010 and each year it got significantly better as peak age and continuity trends improved as well. I had too conservative a projection for the Pirates defense this year because I didn’t weight that trend enough and, while I wouldn’t have had them winning 90 games, I should have had them above .500 because a better defensive projection would have had them scoring more runs than they allowed.
The Pirates finished 94-68, and based on my deeper “cluster luck” calculations I have them as a 92-win team overall. In other words given their exact level of production hitting and pitching, Pittsburgh is a 92-win team in 94-win clothing.
Ever wonder if a personal shopper is a good investment? Well the Cardinals are wearing 97-win clothing and everyone from Las Vegas to national pundits think they look pretty good in front of the camera. However, I think they have the same 92-win body of work the Pirates do.
How is it possible that a team that outscored the Pirates by 149 runs while allowing only 19 more could possibly grade out to the same level? The answer is “cluster luck.” Simply stated the Cardinals benefitted from extraordinary sequencing all year, especially on offense. I’ve gone over cluster luck before in both my book and in dozens of different blog entries so let it suffice here that the Cardinals hit way too many singles (and not enough home runs) as a ratio of their total hits to have scored 783 runs in a league where only the park-aided Colorado Rockies (706) also scored more than 700.
So while it appears the Pirates are vastly underrated in this series, why am I still picking St. Louis to win the series in rather easy fashion? Three reasons:
1) Look at the Pirates first-half record above. On a 100-win pace halfway through the year, Pittsburgh needed to win its last four games to finish over .500 in the second half. The reason why is simple – the best defense in the league through 81 games, by far, was essentially average the second half of the year.
2) The best way to counteract the Cardinals low-strikeout, high-contact offense which results in a lot of singles is to turn batted balls into outs. The Pirates haven’t done that at an elite level since the end of June.
3) While the actual combination of hits, (including type of hits), walks, etc would, over many iterations, most likely result in both teams finishing with the same records, the Cardinals have a better chance of achieving better results during the series. That’s because the Pirates bullpen had a better ERA by more than half a run (2.89 vs. 3.45) but the Cardinals relief corps actually demonstrated a better set of skills. With the 4th lowest strikeout rate of any bullpen in baseball in 2013, the Pirates 3rd best ERA in the majors is simple unsustainable.
Summary:
The Pirates are the best story in the playoffs and the atmosphere Tuesday night in Pittsburgh lent itself more to a college basketball game than an MLB playoff game. Look for the Pirates to remain feisty throughout this series but suffer a late-inning collapse or two to the relentless Cardinals offense. The Pirates may be underrated, but with Francisco Liriano only available for one start and facing a high-contact offense in St. Louis, the Cardinals post exactly the wrong type of team to exploit that misperception. Cardinals in 4.
National League Divisional Series:
Los Angeles vs. Atlanta
I’ve stated above that both the Pirates and Cardinals performed at the level of 92-win teams during the 2013 season. How do the Dodgers and Braves compare? Very, very similarly. The Dodgers also had the results of a 92-win team which in their case matches their 2013 win total. The Braves won 96, but grade out to a 93-win team.
Therefore, despite having the “best” team over the 2013 season in the National League the Braves are very big underdogs (+135, implied win percentage of 42.6%). They deserve a value purchase. I’m still taking the Dodgers in 5 for the simple reason that they were a 92-win team with the incomparable Clayton Kershaw starting just under 20% of their games this year. In a 5-game series, he’ll start 40% of the games. Dodgers in 5.
Mop Up Duty:
Joe Peta is the author of Trading Bases, A Story About Wall Street, Gambling, and Baseball* (*) Not necessarily in that order, a Dutton Books/Penguin (U.S.A.) publication currently available wherever books are sold. Here are three on-line booksellers you can currently choose from:
http://www.amazon.com/Trading-Bases-Gambling-Baseball-Necessarily/dp/0525953647/
http://www.barnesandnoble.com/w/trading-bases-joe-peta/1111308063?ean=9780525953647
http://www.indiebound.org/book/9780525953647
He is also the author of Trading Bases, the Newsletter, a companion piece to the book. If you have been forwarded this issue and would like to be placed on the mailing list, please send an e-mail to tradingbases@gmail.com
All newsletter archives are located at http://tradingbases.squarespace.com
You can follow me on Twitter here: @MagicRatSF
If you want to be taken off the e-mail list, please let me know at tradingbases@gmail.com
Reader Comments (1)
Regarding the Pirates, do you have any evidence that 82 games worth of defense is predictive? I..e, why would you massively change your view on a team's quality based on a characteristic that is (apparently) not very stable? I think you are pretty strongly implying that the skills underlying defensive performance are simultaneously:
a) Subject to significant change within one season.
b) Stable enough over the short-term to fully respect intra-season change as real and predictive.
I don't think these two premises are very compatible.