So, I decided to look more closely at the Angels' record (I first did it here). As a reminder, I had placed it in the context of Days of Week, Game in Individual Series, and Game of Season Series. This time I am adding runs per game, runs against per game, and Pythagorean record (not to mention the updated record at this time).

Day of Week:

Day          Record   R/G    RA/G   Pythag. Record
Monday       11-3     5.50   2.57   11-3
Tuesday      10-12    4.36   4.50   11-11            
Wednesday    11-11    3.86   3.55   12-10
Thursday      9-6     4.00   4.40    7-8
Friday       15-7     5.45   3.91   15-7  
Saturday     13-11    5.17   3.71   15-8
Sunday       11-11    4.14   4.55   10-12

Game of Individual Series:

Game   Record   R/G    RA/G   Pythag. Record
One    32-13    5.49   3.60   31-14
Two    23-23    4.59   3.87   26-19            
Three  22-21    4.16   4.12   22-21
Four    2-4     2.67   6.17    1-5

Game of Season Series:

Game        Record   R/G    RA/G   Pythag. Record
One         13-5     5.17   3.28   13-5
Two          7-11    3.78   3.39   10-8            
Three        7-11    3.50   4.28    7-11
Four        11-3     6.21   4.71    9-5
Five         8-6     5.21   3.79    9-5
Six          8-6     4.07   4.07    7-7
Seven        5-4     4.33   4.89    4-5
Eight        6-3     3.78   3.33    5-4
Nine         3-5     4.88   4.25    5-3
Ten          5-1     7.17   3.00    5-1
Eleven       2-1     5.67   2.00    3-0
Twelve       1-2     5.67   6.00    1-2
Thirteen     1-2     3.33   7.00    1-2
Fourteen     1-1     2.50   3.50    1-1
Fifteen      2-0     4.00   1.50    2-0

So, why did I do all of this? Well, first of all, I looked at the Day of the Week records because it was weird how inconsistent the Angels have been this season. When I saw that Mondays and Fridays were by far their best days (combined 24-9 at the time, 26-10 now) and I knew that Game 1s are typically played on those days, I thought it might have to do with the Angels getting worse over each series. It seemed like a pretty strong correlation once I looked at the records that they do much better in Game 1s than afterwards, so I also decided to look how they fared by game in each season series and their best games are typically Games 1, 4, and 10...each of these are very often game 1s in the smaller series. Of course, all of this seems to just mean that everything matches up well to the idea that the Angels perform better in the first game of a series. It does indicate, though, that the difference is do to the change in individual series and not season series.

In order to look further into this (to see if there is a "real" difference or if it is just luck) I looked at their Runs Scored and Against in each situation, as well as the parsed Pythagorean Records. To me, it definitely looks like there is something real there (and there is even a Pythag. Record record drop off between Games 2 and 3 now). Basically if you ignore the Game 4s in the Individual Series (the 6 Game 4s means that it is way too few data points, as opposed to the regular small sets of the other numbers) you can see a dropoff in runs scored per game (first by almost a run, then by almost another half of a run) and a concurrent increase in their runs against per game (first by a quarter run, then by another quarter run).

What could this all mean? Well, first of all we need to put in all the typical statistical caveats: small sample sizes, these are correlations not causations, I have no standard deviations, and I have no historical or other team data to even know if this is an interesting set of records, among other things. So, basically, I would like to mention a couple of things that could cause this (any other hypotheses are welcome).

First, I would like to take care of the offensive aspect, since that should be relatively the same from game to game, although not necessarily series to series(whereas the pitching has significant changes game to game but less from series to series). The offense has had a pretty strong drop in its production between Games 1 and 2, and then another decent drop between Games 2 and 3. My thoughts are that this might be do to reputation and coaching. Reputation could be a factor (especially early in the season, but at the very least before they have seen the team) because, before you actually face a team, you do not know exactly what their capabilities are (despite scouting and media reports). This could be strongest against guys like Steve Finley and GA, who, once you have seen them in a game, you might realize are not quite the threats they once were. The one problem I see with this is that it should show up in decreased runs over the entire season (the drop off should happen series to series, rather than just within each series). This could be accounted for by realizing that the team can change a lot between series' and so you have to learn how to beat the "new team" once again. Of course, this does not say much for the team's adaptability (why aren't they the ones that are adapting to teams?). Which leads into my other hypothesis, the coaching problem. I tend to like Scioscia's coaching, but this year has easily been his worst, in my opinion. My thought is that he has been too predictable in what he does from game to game (always using the contact play, etc.). This could even be a possible new metric to look at coaching (I know that people have tried using 1-run games and not really found anything). What about looking at how a manager's team does in similar situations (batting over a series, either season or individual)?

The other thing to look at is pitching, which has also gotten worse over the couple of games in the series' (although not as bad as the hitting). I do not really have any ideas on this, as pitching does change significantly from game to game because of different starters, but it should not change as much over the season (which should average out the differences). My thought is that if the starters and bullpen players were parsed out of this, you might be able to find some difference (maybe the good starters play more Game 1s, maybe the bullpen gets worse by as they see each team more).