Fluke Watch (or really, Breakout Watch): Madison Bumgarner

The following article was written before Sunday’s great performance by Bumgarner. All numbers are from before that start.

Madison Bumgarner‘s numbers this year are kind of….weird, in a way. On one hand, you have an ERA (4.06 prior to Sunday’s start) that is almost certainly the result of some bad luck (a high strand rate and probably high BABIP). Certainly his FIP (2.89 before Sunday) and xFIP (3.47 before Sunday) would indicate that he is due to have that number improve by the end of the year.

On the other hand, Bumgarner’s FIP and xFIP are misleading. Taken at face value, they’d indicate that Bumgarner has improved from last year by a decent bit, to the point where he’s a really good pitcher.

Of course, you might also notice something fishy about those numbers: While Bumgarner’s strikeout rate was a tiny bit better than last year’s rate, his walk rate was definitely worse (by a greater amount than his K rate), and Bumgarner’s groundball rate is also essentially identical to last year’s.

So why does he have improved FIP and xFIP numbers? Well, for one, he’s not giving up home runs at all, which is probably the result of luck or random vvariance. (No, it’s not due to home/road issues, as most of his innings have come on the road, rather than in his pitcher-friendly home park.). That explains his lower FIP. However, what about his xFIP? Shouldn’t that account for this?

Well, his xFIP is also misleading as a result of odd batted ball classifications. See, Bumgarner’s GB rate hasn’t increased at all, which would explain a drop in xFIP, as it would expect fewer homers.

But what has dropped is the number of balls being called “fly balls.” What’s happened is that the loss in fly balls has been entirely taken up by an increase in “line drives.” Thus, xFIP thinks, “Hey he’s reduced his fly balls, so he’ll give up fewer HR!” But really, this is unsustainable, and probably just a result of a combination of random variance/luck and the uneven way that batted balls are classified.

But despite the fact that Bumgarner’s FIP and xFIP improvements are misleading, I think he’s a strong candidate to improve his numbers, including his peripherals, and someone you might think of picking up for your fantasy team.

Bumgarner’s pitches

Bumgarner has five pitches:
A four-seam fastball which seems to be sort of straight in terms of horizontal break with okay rise, but decent velocity for a lefty;
A two-seam fastball with decent but not a lot of tail, and really not much sinking action, either;
A cutter or slider (Bumgarner calls it a cutter, but the pitch is in-between) with cutter-like velocity but movement that is really good for a cutter and more similar to a slider (good cut, lot of sink relative to the fastball);
A change-up with similar movement to the two-seam fastball, but more sink and around 83-84 MPH;
A curveball with average velocity and not much drop or great horizontal action (a slurvy pitch except in speed).

Bumgarner, a lefty, throws all of these pitches from seemingly the first base-side of the rubber on seemingly the very edge of the rubber, using a delivery that isn’t really sidearm, but is slightly low and more horizontal than most pitcher’s release.

The end result is that his four-seam fastball (and two-seamer and change-up), despite not having a lot of horizontal break, crosses the plate on a very sharp angle. In other words, to the batter the pitch appears to be far from straight, which should add to the pitch’s effectiveness. (Oddly enough, this effect causes his slider/cutter to be the pitch that approaches the plate the most head-on/straight of any of his pitches, which is the reverse of what normally happens with a cutter.)

Now, Bumgarner’s pitches have changed over the last year: Each of Bumgarner’s pitches has gotten 1-2 MPH FASTER over the last year. The movement on said pitches has basically remained unchanged, but the velocity change would appear to be real. (It’s not simply an effect of a hot gun at AT&T park, as the increase remains on the road).

As a lefty, Bumgarner faces righties roughly 75 percent of the time, so getting these batters out is most crucial for him. And in reality, only two pitches are used significantly by Bumgarner to do so, his four-seamer and his slider/cutter (from here on in, I’m just going to refer to this pitch as a “slider”). The other pitches are there, for sure, but those two pitches make up the bulk of his work, and contain the reason why a breakout may be coming for Bumgarner.

The usage of Bumgarner’s pitches

In 2010, Bumgarner relied heavily upon his four-seamer to do his work,with the other three off-speed/breaking pitches being used near equally the rest of the time. But when the count grew worse for him, Bumgarner would really begin to rely upon the fastball.

The Incompleat Starting Pitcher
The end of the nine-inning start and how we got here.

In 2011, Bumgarner’s usage of pitches has changed. Now, Bumgarner’s slider has taken a much greater importance, clearly being his second option after his four-seamer. In addition, Bumgarner is relying upon his slider even in worse counts; in fact, on 2-1 counts, Bumgarner uses the slider MORE OFTEN than the fastball! The pitch is used over 20 percent of the time on every count except for 3-0.

This increased usage of the slider has come at a decrease in Bumgarner’s usage of the change-up mostly, but also at the expense of his fastball.

Really, this is an odd development if we consider the pitch as a slider. Sliders are generally used as weapons against same-handed batters, with the change-up being the pitch used against opposite-handed batters. But Bumgarner’s change-up usage to righties (opposite-handed batters) has decreased in favor of the slider, which is quite odd from that perspective.

This is not the only respect in which Bumgarner’s usage of his slider has become more cutter-like. In 2010, the pitch was located like a breaking pitch—mostly at the inside and low corner to right-handed batters—and more often than not, the pitch was located out of the strike zone entirely.

In 2011, Bumgarner is hitting the strike zone over 10 percent more often (52.2 percent compared to 41.4 percent) and is locating his pitches more in the middle-low part of the strike zone, though still with an inside bias. This is more like what we’d expect from a cutter, which is used by many pitches similarly to a fastball.

Bumgarner’s results

Bumgarner’s results this year on his slider/cutter have, oddly enough, actually seemed to get less impressive. His swinging strike rate is down from 12 percent to 10.5 percent, while his GB rate on the pitch is also down four percent.

However, these numbers obscure one key improvement in Bumgarner’s slider: the pitch is being called for a ball much less often, down from 33.0 percent last year to 27.1 percent this year. The end result is a more effective pitch overall.

By contrast, the opposite result has occurred in Bumgarner’s fastball this year: The pitch is being called for a ball roughly three percent more often (a significant amount given the pitch is used over half of the time) and has had a worse GB rate, but has had the pitch’s swinging strike rate essentially DOUBLE.

By expected run values, this is, in fact, resulting in the pitch being less effective than last year, though not as much as the slider’s run value has improved.

Bumgarner’s overall usage of this pitch has really not changed—he’s locating the pitch in the same area of the zone as in 2010, hitting the strike zone equally as often, but has gotten this change in results. I suspect the cause of these results has been the increased velocity on the pitch and part of the impact of Bumgarner’s increased slider usage.

These results are what give me great hope for Bumgarner. The increased usage of his slider and change-up in its location has resulted in Bumgarner overall having his swinging strike rate increase from 7.8 percent in 2010 to 8.9 percent in 2011, while Bumgarner’s rate of hitting the strike zone has essentially stayed the same (actually, it’s increased slightly).

This is a significant change, and it’s one we’d expect to result in Bumgarner having a greatly improved K rate alongside a similar BB rate to 2010. Thus, an improvement in his peripherals would seem a likely possibility for Bumgarner.


The last two “Fluke Watch” posts have involved the idea that sometimes pitcher improvements in their peripherals are unlikely to continue because the pitcher’s pitches have remained the same. With Bumgarner, the opposite is true: His pitches seem to have improved, as has his usage of them, and so we might expect to see his peripherals rise throughout the year. As a result, he’s certainly a good candidate for a breakout.

*Note: This article was written before Bumgarner dominated the Cleveland Indians with 11 strikeouts to one walk. I wouldn’t necessarily expect that result every time out, but it’s a good example of what I’m talking about regarding the future of Bumgarner.

Print This Post
Sort by:   newest | oldest | most voted
They did not limit him last season when he blasted past his prior IP highs by a lot. Great article analyzing Madbum’s pitches.  Very interesting! I wonder how much of the changes noted relate to him having a new catcher?  Also, I would note that ATT has not been a pitchers park for a number of years now, most has it as slightly in the hitters favor now.  I think the addition of Petco shifted the road environment downward, as there are still obvious things that help pitchers there, like RF being hard for HRs. And, really, his ERA was… Read more »
Charlie Lorber
Charlie Lorber

I picked him up in a trade a few weeks ago. We have a k/bb category as well as innings.  Do you think the Giants will limit his innings as we get into the later part of the season?