By “run efficiency” I assume you mean outperforming the Pythagorean Record. (Hardball Times does list one NL team with a better Pythag overperformance than the Astros: the Padres.)

Just a cursory glance at the Hardball Times’ close games records would indicate a relationship between records in close games and Pythag differential. If one were looking at overperforming in close games, I would compare the actual record in close games to what the record would have been in close games if the winning % was equal to the team’s overall winning %. Call this “close vs. expected.”

Astros +6 Pythag Variance +4 Close Games +5.2 Close vs. Expected
Mariners +10 Pythag Variance + 16 Close Games +5.8 Close vs. Expected
Padres +8 Pythag Variance + 9 Close Games +7.2 Close vs.Expected

Among the top overperforming Pythag teams, the close vs. expected margin accounts for a large proportion of the Pythag variance.

As I have said previously, I think studying the Angels would yield some information on some team construction factors which lead to pythag overperformance. I say this because the Angels are the serial Pythag overperformers, having overperformed Pythag for something like 7 consecutive years, if I recall. When I get a chance, I will post some more on this subject, because it always intrigues me.

But I think the numbers indicate that we should be looking at factors which affect close game performance if we want to explain the Astros, Padres, and Mariners performance.

Okay, I was a history major, so help walk me through this one.

If the Astros were above average in Run Efficiency, does that mean they were great at scoring runs, or just great at converting whatever scoring opportunities they had into runs? In that case, they may have had 200 less opportunities than, say, they Cubs, but managed to convert a higher percentage into actual runs, which means they were more efficient, but not more prolific. Does that make sense?

I understand the pythag records and all that, it’s just the efficiency bit that gives me pause. What are we really looking at here?

The Astros, in 2009, were better in Medium and High leverage situations than in low leverage by a fairly decent amount too. So it seems, to me, that they managed to make some their RS count for a lot more than just the raw number would indicate. They posted their highest OBP in high leverage situations.

Picking up my earlier statement about the serial overperformers, the Angels. I find it interesting that Ed Wade said he wanted to construct an offense like the Angels when he first came to the Astros. Some typical characteristics of Angels’ teams: high batting average, high contact hitters; good base running and speed; high batting average with RISP. Typically the Angels are near the bottom of the AL in HR ranking, even though they were much better at hitting HRs this year. Despite the analysts who who view clutch hitting as a random variable, the Angels have consistently hit well with RISP in recent years. (Each year, we see some sabermetric writer predict a collapse by the Angels because their success is driven by “fluke” clutch hitting..but it never happens. The Angels are No. 1 in the majors in hitting with RISP this year.)

Now I’m definitely in the realm of hypothesis, rather than proof or conclusions. Maybe teams predicated on contact hitting, high batting averages, and batting averages w/RISP, but relatively low power, can overperform Pythag, because these characteristics are more beneficial in close games, but tend to produce few blow out wins.