BP Comment Quick Links
![]() | |
March 31, 2011 Between The NumbersProjected Standings and Actual PerformanceI decided to take off work for my day job tomorrow for a religious holiday. Yes, I belong to the Church of Baseball—how can I be asked to work on a day like tomorrow? Tomorrow is a time to focus on the important things in life, and for me, this is a mixture of baseball and its statistics. It’s a time to finally watch nine men stroll out onto green grass and officially move the baseline of the expectations that my colleagues at Baseball Prospectus and I have been zeroing in on since the Giants ran on to the field to celebrate their 2010 World Championship. With the release of our final version of the PECOTA-Projected Standings on the Depth Charts page for 2011, I am reminded of an article that I wrote last September called Sabermetric Teams and Sabermetric Scouting. In that piece, I discussed how teams have done as compared with their projected standings over the past five years. My most interesting finding was that teams that were perceived by Baseball Prospectus Staff to incorporate more from sabermetrics into their front office’s decision-making tended to under-perform their PECOTA-projected records. This was interesting, because it suggested the importance of contextualizing what statistics suggest, and even Major League Baseball clubs miss the same information that our analysts miss as well. However, these sabermetrically inclined teams also fared better overall relative to clubs with similar payrolls that were less sabermetrically inclined, which also highlights the importance of listening to what analysts like us have to say! Interestingly, the correlation between teams over-performing their projections in 2005-09 and their over-performance in 2010 was a shockingly high 0.43, meaning that PECOTA’s limits are similar each year. With that in mind, compare projected records from this year’s Depth Charts with the past five years’ over-performance.
In the American League East, we have the Orioles projected eight games ahead of the Blue Jays, but history suggests that those two teams might belong closer together. Never count out the White Sox or the Angels either. These two teams that tend to elude PECOTA as well. Of course, even our less than rosy Indians projection this year will be too high, if history repeats itself. In the National League East, watch out for the Marlins, who have beat their projected records by 5.6 wins per year over the last five years and are projected to finish only four games behind the Phillies. On the other hand, the Nationals might be in line for another high draft pick to supplement their recent top selections. Pirates fans have been excited to see their squad projected out of the cellar for the first time in years, but the Pirates have historically fallen 8.2 wins short of where we project them…which is right about where we have them now. What our Depth Charts do is provide a baseline—an unbiased, quantitative assessment of what a team’s true talent level is. However, it is important to compare this to how teams have exceeded our expectations in the past, and get a more complete picture of what to expect as a new season begins! Happy holiday!
Matt Swartz is an author of Baseball Prospectus.
|
Matt-
It seems to me that this should almost be broken down into two categories if possible. The first would be luck or unquantifiable intangibles (over/under performing the team's pythagorean expected record). The second would be injuries or trades that weren't accounted for in the team's preseason projections. Is there a way that you could use actual WAR or a similar stat to isolate the intangibles from the midseason changes?
That's a good idea. I think the two categories you mention are important, but the unquantifiable intangibles include real skills too. I think things like coaching, synergy, etc., are all part of that. Once the new WARP statistic is available for all years, I'm going to be able to run that a little easier (and I plan to), and I can also check out things I do like the No Turnover Standings for last year too in which I follow the same sum-of-all-WARPs methodology you're talking about.