If newly acquired starting pitcher Jason Jennings can repeat something close to his 2006 season in 2007, then the Astros will have found a counterpart for Roy Oswalt in the starting rotation comparable to what Carlos Lee adds to Lance Berkman in the batting order.

In 2006, Jennings ranked 12th in the National League with a 3.78 ERA. He also finished 15th with a .258 batting average allowed, 21st with a .329 OBP allowed and ninth with a .386 slugging percentage allowed.

In fact, although Oswalt led the league with a 2.98 ERA, eight-tenths of an earned run better than Jennings, they were not so different when it came to some other categories:

Pitcher Avg OBP Slg OPS -------------------------------- Oswalt .263 .299 .403 .702 Jennings .258 .329 .386 .715

While Jennings held opponents to fewer hits per at-bat, he issued a lot more walks, but he also held them to fewer total bases per at-bat. Here?s the same breakdown on a per-nine-innings basis:

Pitcher H/9 BB/9 BR/9 2B/9 3B/9 HR/9 TB/9 ------------------------------------------------- Oswalt 9.0 1.6 10.8 2.24 0.16 0.73 13.7 Jennings 8.8 3.6 12.5 1.53 0.34 0.72 13.1

Jennings allowed fewer hits, fewer home runs and fewer total bases. He did walk two more batters per nine innings, and perhaps that explains the considerable difference in ERA. But maybe not all the difference.

Jennings had a similar discrepancy in the split between his statistics at home and away in 2006:

Split H/9 BB/9 BR/9 2B/9 3B/9 HR/9 TB/9 Avg OBP Slg OPS ---------------------------------------------------------------------- Home 8.8 4.0 12.8 0.98 0.27 0.80 12.7 .262 .339 .378 .717 Away 8.7 3.2 12.2 2.03 0.41 0.65 13.5 .254 .320 .393 .713

Jennings was better at keeping runners off base and surrendering the long ball on the road, although he got tagged for fewer doubles and triples, and hence fewer total bases, at home. Summing all that up, his home OPS and road OPS were virtually identical.

But his ERA was better at home by half a run, 3.56 to 3.97. These are not the only examples of a disconnect between Jennings’ ERA and his peripheral pitching statistics.

Although Jennings had an outstanding 2006, the previous season he yielded a 5.02 ERA. Part of that is explained by pitching precisely half his starts and innings at Coors Field prior to the humidor. Jennings had 10 games started and 61 innings pitched both at home and away in 2005.

His home and away splits in 2005 were similarly quirky to those in 2006:

Split H/9 BB/9 BR/9 2B/9 3B/9 HR/9 TB/9 Avg OBP Slg OPS ----------------------------------------------------------------------- Home 11.2 4.6 16.1 2.51 0.00 0.74 15.9 .308 .388 .437 .825 Away 8.0 4.6 13.0 2.21 0.15 0.89 13.1 .237 .333 .390 .723

With that 102-point difference in OPS, you might think Jennings allowed far fewer earned runs on the road than at home. The actual difference was just two earned runs, for a 5.16 ERA at home and a 4.87 ERA away. In other words, at home he allowed 21 more hits, 19 more total bases, the same number of walks and one fewer hit batsmen, but surrendered just two more earned runs.

What’s the point of raising this? For one thing, it may indicate that Jennings really was about as good a pitcher in 2005 as he was in 2006, and that may mean that the chances are better that he can turn in a comparable performance in 2007.

Comparing his away statistics for the past two seasons, it’s not apparent that Jennings really was that much different in 2006 than he was in 2005 away from Coors Field:

Split H/9 BB/9 BR/9 2B/9 3B/9 HR/9 TB/9 Avg OBP Slg OPS ----------------------------------------------------------------------- 2005 8.0 4.6 13.0 2.21 0.15 0.89 13.1 .237 .333 .390 .723 2006 8.7 3.2 12.2 2.03 0.41 0.65 13.5 .254 .320 .393 .713

Jennings had a 4.87 road ERA in 2005 in contrast to his 3.97 road ERA in 2006. That huge difference in ERA masks the fact that in 2005 on the road, Jennings allowed fewer hits and total bases, although he did walk more batters, than in 2006 on the road.

There are a few reasons why a pitcher’s earned runs allowed might not synchronize with his hits, total bases and walks allowed.

First, if you have two pitchers yielding 10 baserunners per game, the pitcher who tends to scatter baserunners throughout the game should be more likely to have a lower ERA than the pitcher who has the habit of allowing baserunners in bunches. Second, a starting pitcher’s ERA may be affected by the performance of his bullpen in stranding inherited runners or permitting them to score.

Third, while a pitcher’s ERA doesn’t reflect events that contribute to unearned runs, his OPS allowed will incorporate every hit, walk and total base surrendered by the pitcher. Finally, there’s simply random chance, or luck. There’s no statistical rule that says the same number of baserunners and total bases, even when equally distributed, must result in the same number of earned runs.

Just as a team’s OPS is closely correlated to its runs scored, a pitcher’s OPS allowed bears a tight relationship with his ERA. For all starting pitchers with at least 162 innings pitched in a season from 2000 to 2006 (a sample of 588 pitcher seasons), ERA’s correlation was 0.75 with batting average, 0.78 with OBP, 0.84 with slugging percentage and 0.89 with OPS. A perfect correlation is 1.00, so the 0.89 correlation between ERA and OPS allowed is a very strong relationship.

From that same sample of pitchers, 37 of them allowed an OPS between .713 (Jennings’ 2006 road OPS) and .723 (Jennings? 2005 road OPS). Their ERAs ranged from 3.09 to 4.47. For the group, the average OPS was .717, while the average ERA was 3.86. So Jennings’ 3.97 road ERA in 2006 is a little bit higher than what one might expect given his .713 OPS allowed, and his 4.87 road ERA in 2005 was a lot higher than what one might expect given his .723 OPS allowed.

Moreover, if you take the 50 pitchers in the sample group with an ERA between 3.87 and 4.07, *i.e.*, clustered around Jennings’ 2006 road ERA of 3.97, only 15 of them were better than Jennings’ .713 road OPS allowed in 2006, and the average OPS for the group was .733.

Even more drastic, if you take the 41 pitchers in the sample group with an ERA between 4.77 and 4.97, *i.e.*, clustered around Jennings’ 2005 road ERA of 4.87, none of them was better than Jennings’ .723 road OPS allowed in 2005, and the averages for the group are nowhere near as good as Jennings’ peripheral numbers:

Split ERA H/9 BB/9 BR/9 2B/9 3B/9 HR/9 TB/9 Avg OBP Slg OPS ------------------------------------------------------------------------------- Jennings 4.87 8.0 4.6 13.0 2.21 0.15 0.89 13.1 .237 .333 .390 .723 Average 4.86 9.7 3.0 13.1 1.98 0.24 1.24 15.9 .275 .337 .451 .788

Other than ERA (and walks), Jennings’s statistics were distinctly better than the sample group’s average. It seems pretty unlikely that a pitcher posting the kinds of numbers Jennings did would have an ERA as high as the sample group’s average.

Which doesn’t mean the same thing can’t happen again. Sometimes lightning strikes twice or even three or four times. But the discrepancies do suggest that Jennings’ 2005 performance on the road was more akin to the fine numbers he produced in 2006 on the road and in a humidor-affected Coors Field than his ERAs indicate. Simply put, Jennings was possibly a better pitcher in 2005 than you think, somewhere close to what he was in 2006.

Maybe it’s just wishful thinking, but here’s to hoping that translates into a successful performance by Jennings for the 2007 Astros.