johngordonma was asking about record run-differential overachievers. I couldn't find a list, but since it's an off-day, I used the time to scope out the records in this area. I went as far back as 1969, when baseball went over to a divisional format. The lists below show the year, the team, how many games they were above their expected Pythagorean record, and the final, actual record they posted. The cut-off point for inclusion is +10, the same as the current Diamondbacks team.
- 2007 Diamondbacks: +10 (63-50)
- 2005 Diamondbacks: +11 (77-85)
- 2004 Yankees: +13 (101-61)
- 1997 Giants: +11 (90-72)
- 1984 Mets: +13 (90-72)
- 1981 Reds: +10 (66-42)
- 1978 Reds: +10 (92-69)
- 1977 Orioles: +10 (97-64)
- 1974 Padres: +10 (60-102)
- 1972 Mets: +11 (83-73)
- 1970 Reds: +11 (102-60)
So while the Diamondbacks' 'luck' is extreme, it's certainly not unprecedented, with 11 teams in 39 seasons posting the same or better marks. The 1997 Giants are of particular interest, since they made it to the playoffs, despite allowing more runs than they scored (784-793). Of course, that is easier with the steady expansion of the playoffs, but the Giants were their division winners, two games ahead of the Dodgers.
Their "luck" was almost all in the first half, when they were 51-36, despite having a differential of -18. They were good, rather than spectacular, in one-run games (23-17), but great in extra innings, where they won eleven of fourteen. Like AZ, that first-half record was padded by blowouts: they lost four times by ten runs or more, including 19-3 to the Expos, and never won by more than eight. [As an aside, for a home series against the Phillies in April, the Giants averaged crowds of just 7,134. What a difference some flaxseed oil can make, eh?]
I also notice the almost total lack of American League teams on the list, with only two teams having reached +10 in almost forty years. Is run differential more accurate in the American League, with the DH? That would play into the theory that managers make a difference, with the Senior Circuit giving more opportunities for pinch-hitting, sacrifices and other factors that can lead to wins, without necessary improving run differential much. Teams are probably also more likely to play for one run, rather than sitting back and waiting for your DH to jack one out of there.
The good thing is, run differential has no memory. The fact that Arizona is currently +10, doesn't mean they will be -10 the rest of the way. It's like a roulette wheel, where past results will have absolutely no impact on future ones. That is, of course, if you subscribe to the theory that it is purely luck. If, on the other hand, it is all down to the managerial genius of Bob Melvin (or the brand of aftershave favored by Chad Tracy, or the Flying Spaghetti Monster - basically, anything apart from luck), then there's no reason we can't keep on outperforming Pythagorean expectation. Just as long as we keep winning the close games and losing the blowouts.
And that's exactly what we've been doing, all year long. Here's a breakdown of the team's real record, their run differential, and their expected record based off that, on a month-by-month basis [August's games have been included in the July figures]:
W L RS RA EW EL
April 16 11 116 108 14 13 +2
May 16 12 123 116 15 11 +1
June 14 13 108 127 11 16 +3
July+ 17 14 120 145 13 18 +4
As you can see, the team has been consistently "lucky" throughout the entire season. Well, the degree of luck may have varied, but it's always been on the good side. And the team has been consistent, playing .500 or better ball every month to date. That's something not even the World Series team in 2001 could say, as they had losing records in both July and September. This July was particularly striking: we started off 4-12, then roared back to win nine of the last ten and end with an even record. August is starting off quite well too, 4-1 thus far.
Okay, stuff to do, so I'll stop here. Fantasy Baseball update to follow, though my enthusiasm for that is understandably hard to locate this week! And I've also been working on SnakePit 1.0.1, so I should get that finished off tomorrow morning. Hope you're enjoying the off-day as much as I am.