FanPost

Losing the WAR and Euclid's Revenge

I am combining two fanpost ideas I’ve had in one post, and hope all of you can follow. There is some relationship between these ideas, unfortunately.

Part 1: Losing the WAR

This part of the fanpost is a follow up, at approximately the 2/3rds point of the season, to my pre-season analysis about the possibilities of this Phils team from an fWAR point of view. In short, we just could say LOL to all of my analysis. But I think it might be useful and instructive to see exactly where I was wrong, to see exactly how they (and I) failed. As you may recall, my analysis was to see how well the Phils would do if they performed up to their peak fWAR from the 2011-13 seasons. If they would play close to their 2011-13 peak, they would clearly compete.

Note that the 2014 amounts below are what would be projected from here at the end of the season if they perform at the same level during the last 1/3 as the first 2/3rds of the season.

Background: in the original article I asserted that J-roll especially, but also Marlon and Chooch, would drive the Phils’ fortunes this year. The story turned out to be more complex.

Player/ 2011-13 peak/2014 projected actual

A. Meeting or exceeding expectations

Chase Utley 3.9/ 4.5 (swoon!)

Chase is the only Phil projected to meet or exceed his 2011-13 peak. And even that is not guaranteed to happen as he is still riding the wave from that tremendous start. Normally when teams win, several players perform at or near their career peak levels (e.g., Rollins and Utley in 2007-08). Here, players are not even performing to their recent levels of success, let alone career peaks.


B. Close to Expectations (within 2 WAR of 2011-13 peak)


Jimmy Rollins 4.7/3.6
Cole Hamels 4.2/3.6
Antonio Bastardo 0.9/0.3
Roberto Hernandez (!) 0.9/-0.1
Kyle Kendrick 1.7/0.6
Jonathan Papelbon 3.2/1.8
Ryan Howard (!) 1.5/0.1
Ben Revere 3.0/1.2
Cody Asche 2.0/0.0


What is interesting about this list is that most people in this grouping aren’t "that" far off expectations, but collectively this represents about 12 less team WAR. Only Cole Hamels, given his recent performance, might move from this category into group "A" by the end of the season.

Oh, you are surprised about Howard and Hernandez being in this category and not further down? Well, in the original article I said that neither of these players would impact the Phils’s fortunes very much. Howard’s 2011-13 peak was 1.5 WAR, and I said in the earlier piece that I expected that he would almost certainly be within 1 WAR of that projection because if he wasn’t, Ruf would play. I think my analysis would have been very close had Ruf not gotten hurt. The difference between my pre-season expectations and the Phils management’s is that I did not have any expectations about Howard.

Curiously, Revere would actually be close to his 2011-13 peak if his defense was equivalent to his 2011-13 peak defense. Offensively he’s nearly what we would have projected. Asche’s WAR was a "placeholder" since this was his first full year, but the fact that he so far projects out to be a perfect replacement player is not heartening.


C. Below Expectations (within 2-3 WAR of 2011-13 peak)

Marlon Byrd 4.1/2.0
A.J. Burnett 4.0/1.7
Diekman/Rosenberg/Giles/DeFratus/Hollands 2.0/~0.3
Bench 2.0/~-0.2
Carlos Ruiz 5.2/2.5
There are caveats in this category. Marlon’s decline comes mostly from a negative fielding assessment, which as of today is less negative per UZR/150 than it was earlier in the season. If he stays with the Phils he almost certainly will move into the "B" category above. Chooch’s decline is greatly impacted by injury, so we can expect him to head to the "B" category by the end of the year. Burnett can blame his hernia. I guess.

The bench has been even worse than we thought it was going to be. Joe Catz’s analysis that a bad bench would seriously impact the team’s performance, given that there would be no one to "take up the slack", proves prescient.

You knew the non-Pap bullpen was bad. The only reason it is even this good is because I’ve included Giles who wasn’t in the original analysis.


D. The basement, at least relatively speaking (3 or more WAR less than 2011-13 peak)

Cliff Lee (!) 6.5/2.4
Dom Brown 1.7/-2.0 (!)

We have heard about Dom’s problems. But even though at 2.4 Lee himself is far from "dreck", that is a huge difference from his peak, and right now it isn’t clear he is going to even reach that 2.4 mark if he does not improve from current performance.

Overall, we can see that this disaster has been a team effort. I suggested that it would be J-roll’s performance that would stir the drink, or not, along with Chooch and Byrd, and that’s clearly not true. J-roll is almost as good as we ever could expect, and while Byrd and Chooch have not performed to expectations, there is plenty of blame to spread around, and not as much on those two. Other than Lee and Chooch, and to some extent Cole, they’ve been mostly healthy while performing terribly.

So, I suggested that at their peak 2011-13 values, Phils players would combine for 49.8 WAR: 97 wins!. An amount even close to that would yield a competitive season, at least.

In fact, the total from the above is about 21 WAR. That would play out to 47.7 plus 21, or approximately 69 wins. Which brings us to part 2.

Part 2: Euclid’s Revenge

Many of you are familiar with the oddly and inaccurately named Pythagorean theorem of baseball, where the percentage of runs scored vs. runs allowed is translated into their related percentage of wins and losses. This is not one of the statgeeks’s favorite stats since it is a blunt instrument.

In fact, at about the time I created the preseason fanpost I looked through several of the teams’ WAR at the end of 2013, in part because I noticed that there was a bit of variance between team WAR and performance, especially for teams like St. Louis. But I then noticed that the Phils outperformed both WAR and Pythagorean by a significant amount; both of these measures would have seen the Phils win somewhere between 63-66 wins, depending on the measure.

So, just as an exercise, I looked to see the trends in the Phils runs scored and Pythagorean measures over time. I was a bit startled to see the relationship.

Year/ Pythag projection/ Actual wins/Runs scored/Difference between Pythag and actual

2013 66 73 610 +7
2012 81 81 684 0
2011 103 102 713 -1
2010 95 97 772 +2
2009 92 93 820 +1
2008 93 92 799 -1
2007 87 89 892 +2
2006 86 85 865 -1
2005 89 88 807 -1
2004 86 85 840 -1
2003 90 86 791 -4
2002 79 80 710 +1
2001 84 86 746 +2


That’s pretty amazing to me. Other than the 2013 outlier, the actual numbers are very, very close to the Pythagorean. When I first looked at this, I thought LOL Bowa for 2003, but in fact, overall his Pythag from 2001 to 2004 is about -2. And his worst year relative to projections is -4, but if you look at SFG, Bruce Bochy has *multiple* years of both +6 and -6 relative to Pythag projections. Tony La Russa was all over the map with the Cardinals, with years of +7 and -7. It’s not clear to me what this says about "managerial performance vs. luck", if anything.

But Charlie’s was spookily close. Over his entire tenure, never more than 2 games off from Pythag, and overall from 2005-2012 he was exactly +1 relative to Pythagorean projections.

And in 2014, the Pythagorean would yield something like 71 wins, close to the projected 69 wins via WAR above with around 610 runs scored once again.

Here are some conclusions I would draw:

1. Overall, scoring matters. With the exception of 2011, the Phils were more competitive when scoring more runs, and less competitive when they didn’t. I know that sounds like something from the John Madden School of Insight, but the numbers are pretty striking. They do not have the ability to put up runs now, and it means that pitching must be near-perfect. That downward trend in runs scored is clear and a big problem.


2. The scoring decline did not begin in 2012 but 2011. The HOF-type of rotation that the Phils had in 2011 masked a huge percentage decline in runs scored, as they scored the fewest number of runs 2011 since 2002. Once Halladay broke down and Oswalt was let go and retired, this was not sustainable. Having Tony Gwynn as a pinch hitter in 2014 only extended the trend.


3. We were all excited about the Lee signing, but now I have some questions about this, at least with regard to the way the contract came about. I’ve mentioned before that without the Howard early signing, the Phils might have been able to re-up Werth early, and possibly never let Lee go nor acquire Oswalt. But here, you have to wonder if at least some sort of competitive offer should have been made to Werth instead. In particular, the back-ending of Lee’s contract is hurting the Phils when it isn’t clear or not if he is in the twilight of his career; we saw Roy H. descend quickly, as we saw Carlton do the same. At the very least, paying Lee less money up front and more later never made sense from a strategy sense, even if it seemed to make sense to RAJ in a short-term budgeting sense. And remember, if Werth is re-signed sometime in 2010 prior to free agency Hunter Pence is never mentioned in a TGP blog; fyi, the real problem with the Werth contract is the back end, since his WAR/salary ratio so far isn’t that bad. In any case, in retrospect the 2011 "all-in" bet on pitching is proving to be problematic for the Phils in the long run. The Papelbon signing is only the worst choice among several risky choices.


4. If Byrd is traded, this will exacerbate a continuing trend. This makes me nervous. Clearly, if all of the above is true, it’s Burnett and Bastardo that have to go, (unless RAJ is overwhelmed for Hamels or Lee) and any money saved designated for offense. We have to hope that the Phils sign some international free agents, and that players like Franco, Crawford and even Quinn will come through, in a big way offensively even if not defensively, or it is going to be a long long time before the Phils compete again.