After enjoying some All-Star Game inaction, Chris Carpenter pitched like one of the league’s best on Thursday. The Jekyll-and-Hyde routine continued, as "good Carp" — that is, circa 2005-6, emerged as the big Cardinal righty struck out only a modest share of the 27 batters he faced (six) and, more importantly, walked none. He did allow a solo home run to the Dodgers’ best hitter, Andre Ethier. But even that was with a lead and on a first-pitch: Credit Ethier for going up swinging after Carpenter had thrown first-pitch strikes to seven of the previous 11 Dodger hitters. Oddly enough, Carpenter didn’t miss a bat to a non-pitcher, though, until the sixth inning, when Matt Kemp went down on strikes, two of which were swinging. But one particular stat revealed that maybe it wasn’t the vintage Carpenter: He allowed 10 fly balls. So we’re reluctant to claim that he has regained his old form — the 2010 Carpenter is still very much a hybrid.

Other notes:

• Speaking of Kemp, he waved at four pitches during his final two at-bats, one of which was against Mitchell Boggs. The pitch? Cut fastballs, though Boggs brought his about eight-to-10 miles an hour faster than Carp.
• Albert Pujols had three hits, but his first-inning plate appearance left us wondering about his valor. Down 1-2, he pretended that Clayton Kershaw’s bounced pitch hit his foot. Whether or not it did, the attempt is excusable and perhaps even laudable for a lesser hitter who is down in the count. But it was an odd move for the Great Pujols.
• We’re not as bothered by batters striking out as we once were or as some fans still are today. But Allen Craig’s ability to adjust and put the ball in play on two-strike counts led to a couple of runs, whereas a strikeout would’ve left the team blank. He chopped into a run-scoring forceout in the first, and hit a sac fly in the fifth. There is still something to be said for situational hitting, specifically on two-strike counts against a pitcher who features a career 9.43 K/9.
• The Cardinals in general avoided strikeouts, as Kershaw only had one. Kershaw made it easier by not being able to find the plate; at one point, he had thrown 33 balls to 35 strikes.
• On the other hand, Brendan Ryan didn’t exhibit much discipline in his fourth-inning at-bat. With a 3-0 count, Ryan swung at a high-and-tight pitch and popped out to first base. It’s one thing to swing at that pitch when the guy is dealing or when you’ve got two strikes — it’s another to offer when you’re in the driver’s seat. Ryan has struggled at the plate this year with a .247 wOBA, and though he showed last year that he has the tools to hit (.324 wOBA), he’s going to wind up as simply another Adam Everett if he can’t discipline himself.
• We saw two more examples of Cardinal baserunners behaving badly: Matt Holliday took his arrival at second for granted on Craig’s forceout at third in the third inning and was tagged out. And in the fifth inning, Albert Pujols blatantly disregarded third-base coach Jose Oquendo’s stop sign as he rounded third on Holliday’s single to right. Pujols ultimately stopped about a third of the way down the line, but it was clear that Oquendo’s protestations had no effect and that he was going to do whatever he wanted. When Pujols bats or is on base, Oquendo should just sit down.
• The official scorer ruled the second of the Cardinals runs in the fourth inning as unearned. It was Aaron Miles, who had taken second on Kemp’s errant throw after singling. Miles, therefore, would’ve only been on first without the error when Felipe Lopez doubled with two outs. Miles might’ve scored, but the scorer relied on the "benefit of the doubt to the pitcher" explanation in rule 10.16(f):
  

  Whenever a fielding error occurs, the pitcher shall be given the benefit of the doubt in determining to which bases any runners would have advanced had the fielding of the defensive team been errorless.

• Tony La Russa pinch hit for Colby Rasmus in the eighth with Randy Winn, a pointless decision. With a six-run lead, the benefit of Rasmus getting more experience facing lefthanded pitching outweighed any benefit of an additional baserunner.
• The Busch Stadium scoreboard display of the team’s lineups now highlight the batter due up for each team, a positive change.
• Los Angeles had four repeat-digit uniform numbers in their starting lineup: Manny Ramirez (99), Russell Martin (55), Blake DeWitt and Clayton Kershaw (22). The Cardinals have only two total on their active roster. Can you name them?
• Speaking of uniforms, Martin’s is confusing — his nameplate is "J. Martin," the J being the initial of his mother’s maiden name, which is one of Russell’s five legal names.
• Ramirez made headlines by returning from the DL last night, but he was a non-factor. More accurately, he was a negative factor for the Dodgers, flying out harmlessly three times and committing a Manny-like error, absentmindedly trying to field a hit to left. The Manny Show is a little bit like the Eagles, who toured in St. Louis earlier this summer: Once highly entertaining, but now overplayed.
• Whereas most outfielders gather in centerfield to kibitz during a mid-inning pitching change, Manny left Ethier and Kemp to discuss things by themselves last night. We suppose, though, that that didn’t mean Manny wasn’t having a conversation out there.

Yesterday we looked at the "voting constituencies" — fans, players and manager — responsible for the selection of the National League’s All-Star representatives. Today, we’ll review those groups who selected the Junior Circuit stars.

The All-Stars were:

Wow! The first thing you might notice is how well the AL fans did in picking their starters — every player was in the top three best at his position, and seven of the nine were the best at the position. That meant that the fans picked best of the three consistuencies:

Some more observations:

• The fans who voted for the AL starters either had a clearer choice of candidates, or they simply picked more astutely than the fans who chose the NL starters (fat lot of good that did them!).
• Though his hands were a bit tied by having to select someone from the Orioles, Joe Girardi could’ve — and should’ve — picked Nick Markakis, whose 2.2 WAR is better than Ty Wigginton’s 0.4 (Weaver’s Tantrum readers seem to agree in their poll).
• Overall, Girardi really bombed with his reserve selections. In addition to the Markakis oversight, he bypassed the AL’s best player by WAR in Kevin Youkilis (3.4 WAR) for Paul Konerko (2.1 WAR) and opted for his own Alex Rodriguez (2.2 WAR) when he could’ve had a different Alex, Alex Gonzalez, Blue Jays’ infielder (2.4 WAR).
• But Girardi, a former catcher, was better at picking his reserve pitchers. One of his "worst" selections, Joakim Soria (0.9 WAR), was forced a bit as the Royals’ representative. But Girardi could’ve easily chosen one of his league’s best pitchers in the Royals’ Zack Greinke (2.7 WAR). And Andrew Bailey is a reliable closer but an inexplicable choice when Francisco Liriano, Felix Hernandex and Gavin Floyd were available.
• Had the fans chosen Youkilis, whom Nick Swisher edged as the 34th man, they would’ve scored a nearly perfect 1.3.

So which group did the best of all, counting both AL and NL voters?

Another way to look at the groups is to combine them across leagues:

So it would appear that, when it comes to selecting the best, the players know, er, best. Fans aren’t far behind. Ironically, the men responsible for actually managing the All-Stars appear to be least qualified to select them. That the bigger constituencies — the fans and the players — chose better may be due to the "wisdom of crowds" effect. For whatever reason, though, ceding some of the selection-making to the players would seem to be one of the few wise recent decisions regarding the Midsummer Classic.

With so many ways for players to make it to the All-Star Game nowadays, which voting constituency is best at choosing deserving players? Fans, players and managers all have chances to select players — who are the best talent evaluators?

To find out, we ranked all players in each league by Wins Above Replacement, then assigned a value to each constituency’s pick, based on the best player available at the position. For example, fans had the first pick of players, so to speak, so they had the entire pool to choose from. In the NL, the top first baseman by WAR is Joey Votto (indeed, he’s the best at any position). But fans chose Albert Pujols, who is second in the league in WAR at 1B; fans therefore earned a "2" for their pick.

We assumed the following "draft" order: Fans, players and manager. That is, after the fans were finished picking, we assessed the players’ picks based on the remaining players. When positions were prescribed — on the fan ballot (position-by-position) and on the player ballot, on which players had to select five starting pitchers, three relievers and eight so-called "position" players — we created tiers of available players. For example, on the fan ballot, fans choose three outfielders, so we grouped outfielders in tiers of three players each — if the fans had picked the top three outfielders in the league, they would’ve gotten a score of 1 in each outfield spot (rather than 1, 2 and 3). If a player was 16th-best or beyond, we simply scored the selection a 16 (the number of teams in the NL and therefore the number of players available at each position on the fan ballot). When it came to the managers, we created the tiers based on how they decided to structure their benches. For example, Charlie Manuel chose seven pitchers, so we structured the remaining unpicked pitchers in tiers of seven.

We then averaged each constituency’s scores to determine who is best.

Today, we’ll review the NL picks by constituency; we’ll cover the AL in a subsequent post.

National League

So how did they turn out? The players know how to pick ‘em:

Some observations:

• The NL players were spot-on with their starting-pitcher picks, choosing the top five in WAR. Either they read Fangraphs, or WAR matches up well with real-world experience. Or both.
• If you factor in the manager’s choices to replace Jason Heyward in the outfield and to choose a starting pitcher from the available roster, he does even worse. Of the healthy four reserve outfielders to choose from, Manuel picked Corey Hart, the worst of the lot. And of the starting pitchers, Manuel might’ve given the ball to either Roy Halladay or Josh Johnson, both of whom are superior to Ubaldo Jimenez.
• Fans redeemed themselves with their 34th-man pick of the league’s best player, Joey Votto. Otherwise, they would’ve had a 5.75 rating, about the same as the manager.

As we enter the All-Star Break, let’s review the first “half” of the Cardinals’ season. The Cardinals shot out of the gate, going 18-8 to begin the season. In going 10 games above .500, they build a five-game lead on their nearest central-division competitor. They slumped a bit in early May, but by June 5 had re-summitted the 10-games-above-.500 mark. In the meantime, however, the Reds had caught fire and were only one game behind, as the following graph shows:

After their series-ending win in Houston Sunday, the Cardinals now sit in the same place they were back on April 27 — six games above .500. Yet, because of the Reds’ strong play, the Cardinals now find themselves looking up in the standings.

In addition to the Reds’ streaking, the Cardinals have run into some bad “luck.” They’ve been running a gap between their actual winning percentage and their Pythagorean winning percentage since the beginning of May:

The Cardinals are sending five players — Albert Pujols, Yadier Molina, Adam Wainwright, Chris Carpenter and Matt Holliday — to the All-Star Game this year, the most of any team. Manager Tony La Russa recently remarked in jest that “Five All-Stars should mean a 10- or 15-game lead, if you had a manager who’s worth a hoot.” But does a team’s all-star representation correlate with regular-season success?

Let’s look at the Cardinals’ history to get an idea. In the 77 seasons in which All-Star Games have been played (including this year), the Cardinals have had 266 All-Stars, according to their web site. They’ve had as many as eight players named, as they did in 1943, when they won the pennant, but have also won division titles with as few as one, as they did in 1996 and 2002.

To exclude some margin for error, we used Pythagorean winning percentage rather than actual. Also, since the number of teams in the league has changed over the years, we also attempted to normalize the number of all-stars based on the chance that a team could have one. That is, since it’s theoretically easier to have more All-Stars when fewer teams exist, five All-Stars today, when the NL has 16 teams, is a bigger accomplishment than having five in 1939, when the NL had only eight teams. Neither adjustment is going to make a huge difference, but it’s a little more interesting.

So what did we find? The correlation coefficient between the Cardinals’ Pythagorean winning percentage and their adjusted number of All-Stars was a medium-strong .53 (the actual winning percentage correlation was .55). It’s not altogether surprising, of course. Even though La Russa’s comment was said tongue in cheek, it contains a nugget of truth. Maybe it’s too much to expect a "10- or 15-game lead," but it is reasonable to expect a team with enough talent to send five stars to Anaheim — at least three of whose selections were merit-based — to win. Indeed, the Cardinals have a .581 Pythagorean winning percentage. Their recent failures in Colorado notwithstanding, they’re still a good team, and one whose star power should translate to a winning second-half.