How quickly things change when you are in the heart of a pennant race.

Six starts ago, Pirates pitcher James McDonald was putting up some of the best numbers in baseball — 3rd in the National League in both ERA and baserunners per inning. Six starts ago, Buccos fans were steamed that McDonald was not on the All-Star team, and talking about him as a Cy Young dark horse (myself included). Six starts ago, McDonald was seen as a pitcher that would lead Pittsburgh’s charge to the first postseason in a generation.

In reality, six starts is not a whole lot of time to evaluate a pitcher. One time through a tough lineup can sway the numbers a lot, a few bloop hits could drop in or a pitcher could just come out flat on a given night.

Yet one can not deny that fans are seeing a vastly different James McDonald than they saw a month ago.

It is a pennant race, and patience is at a premium. As the Pirates sit right on the Wild Card bubble heading into Thursday, and it is possible that McDonald could be relegated to the bullpen.

That decision is up to Clint Hurdle. But I wanted to look at what has changed for “J-Mac” over those six starts since the All-Star Break (or ASB). Props go out to Brooks Baseball and FanGraphs for the raw data; I could not have done it without them.

1. Command falls to pre-2012 form – In the first half of the season, McDonald was vastly improved at throwing strikes. Part of it seemed to be the influence of veteran pitcher A.J. Burnett and catcher Rod Barajas. McDonald was aggressive in the zone and it got results. But since the break, his strike rate has dropped below his career rate.

2. Fastball velocity is down – I took a look at McDonald’s fastballs (both his four-seam fastball and two-seam sinker) and found a noticeable drop in average velocity. His fastball offerings are down a full mile per hour since the All-Star Break. Keep in mind that McDonald’s 171-inning season in 2011 was the first time he had ever pitched more than 72 innings in a Major League season. Now he is at 141 innings this year and counting, so this McDonald that averages 92 mph on fastballs will likely remain for the rest of the year.

3. Opponents are getting their bats on his fastball – With McDonald’s drop in fastball velocity, hitters have found it much easier to put the pitch into play.

4. His slider has not been as effective – McDonald used the slider as a knockout in the first half, relying on it especially against right-handed hitters. Since then, it has still been a good pitch, but it is more hittable and not the dominant offering it had been. Perhaps the drop in velocity is a factor?

5. Many long innings – People often measure a pitcher’s stress level in a game by pitch count, and their stress level over a season by innings count. Hell, I just did the latter a couple paragraphs ago! But it is also good to see if he is throwing a lot of high-stress innings, which I measure as innings with 25 pitches or more. Given McDonald’s troubles, he has a lot more of these innings that take a lot of pitches to finish. Before the break, he averaged 15.3 pitchers per inning. Since then, he has averaged 19.3 per inning.

Okay, so all of those charts identify the particular problems McDonald has struggled with over the last month or so. But there are reasons to think that he might improve. In fact, there are some very good reasons to think that McDonald will pitch better over the last several starts of his season (assuming he stays in the rotation).

1. He is still making hitters swing and miss — especially with his curve – One of the main elements of McDonald’s game has been generating whiffs. His stuff is dominant at times, especially when he has his fastball-slider combination working. But since the All-Star break, McDonald has mixed in his curveball more often — and for good reason. Since the break, his curve has gotten hitters to flail almost as well as his slider does. Keep the curves coming.

2. Bad BABIP luck – If you’re not familiar with BABIP, it stands for batting average on balls in play. Many sabermetricians say that unless a pitcher generates a lot of groundballs or a lot of line drive, fluctuations in BABIP are largely due to luck. Specifically, if a pitcher’s season BABIP is lower than his career BABIP, that’s good luck. If it is higher, that is bad luck.

McDonald has been unlucky since the All-Star Break. One big reason this should improve is the Pirates’ quality defense, which has the 3rd-highest rate of balls in play converted to outs in all of baseball.

3. Bad home run luck – Another sabermetrics theory on balls in play: pitchers mostly control giving up fly balls, less so for home runs. So in practice, if his season home-run-to-fly-ball rate is higher than his career rate: bad luck. If it is lower than his career rate: good luck. You can see that, like BABIP, McDonald’s luck has been a tale of two halves.

4. His line drive rate has held steady – Despite McDonald’s fastball and slider being put into play more often, opposing have not exactly smacked him around. This falls in line with our feelings that his recent BABIP is simply bad luck. A pitcher should not have that high of a BABIP unless there are a lot of liners being allowed.

So there you have it. The bad and the good of James McDonald. The referendum on whether he will stay in the rotation could come Thursday in his next start, though it would be wise to be patient and see if his luck turns.