FG+ O/U Game: Madison Bumgarner
This week, we’ll be giving away one FG+ membership per day by playing the FG+ Over/Under Game. The wrinkle on this game is that it’s subjective: we’ll provide a player and a number, and you use the comments section to make your best argument for the over or the under. The RotoGraphs staff will pick a winner every day, and that person will get a free subscription to FG+, which includes 11 full-length fantasy strategy articles, 1100 player caps on the player pages, and ongoing access to the FG+ blog, which features the writing the FanGraphs team provides to ESPN Insider on a weekly basis all year.
For our first over/under game, let’s take a look at Madison Bumgarner.
The over/under number for Bumgarner is 20%. As in, will he strike out more or less than 20% of the batters he faces in 2012?
The 6′ 5″ lanky lefty suddenly upped his swinging strike rate in 2012 despite using the same slate of offerings at about the same velocities. If you had access to FG+ this year, you would be able to read about how best to use his Pitch F/x information to answer this question — thanks to Josh Weinstock. And — thanks to Michael Barr — you’d also have an equation for expected K% that would be derived from his 2011 peripherals.
But since you don’t — and yes, I’m suggesting that current subscribers take a back seat and allow those without your knowledge to compete for this — you’ll have to do the best you can. Please use anything other than the xK% equation for your argument, and make it as long or as short as you like otherwise.
Have at it!
With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.
What in tarnation kind of question is this? It’ll be over, and fuck your cityboy math. I’ll strike these fellers out because I wrangle steers in the off-season to compensate for having a girl’s first name. I throw sows across my tobacco field into my bass pond to make my arm strong, so how hard could throwing a baseball be?
Over..
His advanced stats show that:
He has increased the use of his slider each year with a bump in its velocity.
His control has increased ever year since promotion(declining BB rate each yr).
Stats also show he has gotten over his shoulder issues of 09( avg FB 89.2), with a return to a 91.5 average FB velocity.
I’m going to try to keep this brief. I’m also going to take the OVER.
Let’s start with the easy part. Because Bumgarner walks so few batters (a little over 2 per 9IP), a large number of the batters he faces will either strike out, put the ball in play, or hit a home run. On that latter, point, Bumgarner has suppressed home run rate over his career, and this is actually a fairly common occurrence among Giants pitchers. (Assist to Jesse http://www.fangraphs.com/blogs/index.php/dave-righetti-lord-of-the-hrfb-rate/)
So where we’re at now is that roughly 5.5% of his outcomes will be walks and perhaps 2% will be home runs. That leaves 92.5% of outcomes to be split between BIP and strike outs. Last season, Bumgarner K’d a healthy ~22% so the question becomes – is this sustainable? A 9.2% swinging strike rate is roughly in line with that sort of strike out rate – especially given the high number of outcomes split between BIP and K – but again, will he sustain that swinging strike rate?
A lot of noise has been made about his increased slider usage. According to Brooks Baseball (http://brooksbaseball.net/player_cards/player_card.php?player=518516 – set the year tab to 2011 on the second table), Bumgarner’s slider usage was static for all intents and purposes – 20% in 2011 vs 21% in 2010. What changed was that he used a new pitch – a cutter – 13% of the time. Both the slider and cutter generate whiffs about 13% of the time. Given the success of his cutter, he may/probably should opt to throw it even more often in 2012 – hence buffing his swinging strike rate and, ultimately, his strike out rate.
Given his success with the cutter, let’s imagine he increases his usage to 18% of the time and holds steady with his slider, sinker, curve and change up. That gives us the following rates (roughly).
4 seam – 38% usage, 8.5% whiff
sinker – 10% usage, 7.5% whiff
cutter – 18% usage, 13.3% whiff
slider – 20% usage, 13.4% whiff
curve – 11% usage, 11.4% whiff
change – 4% usage, 9.7% whiff
Total – 101%
Again, this is very rough. Interact those above rates and we get a swinging strike rate of…10.7%. That’s significantly above his 9.2% rate from 2011 and could very well be bullish, but it also leads me to believe that his swinging strike rate shouldn’t get any worse. Strategically, it’s not unreasonable to expect Bumgarner to replace some of his vanilla four seam fastballs with superior cutters. We’re seeing this phenomenon around the league. Even if he doesn’t increase his cutter usage, he should be able to maintain at least his 9.2% swinging strike rate, which should be enough to result in a 20% strike out rate.
For what it’s worth, THT’s Oliver projection system estimates that Bumgarner will strike out about 19.8% of batters faced (“about” because double plays and CS aren’t factored into Olivers TBF, I WAGed 20 which probably isn’t right at all).
Apologies if any language (especially about all those percentages) was unclear as this was drafted rapidly.
And yes, I do believe that was the brief answer. When you pull out those brooks baseball cards, it becomes extremely difficult to be succinct.
Over. Everybody talks about how in regaining the zip on his fastball, Bumgarner reestablished himself as a starter who could miss bats. However, what really separated 2010 from 2011 was the discovery of an extremely potent slider. Yes, in terms of pitch value his fastball went from mashable (-5.4 wFB) to respectable (2.2 wFB) but it was his slider that vaulted into unhittable territory. In 2011 he posted a wSL of 17.7, behind only Clayton Kershaw (22.9 wSL) and Ervin Santana (22.0 wSL). Compared to the -0.3 wSL Bumgarner posted in 2010, there’s no question where the jump from 18.2 K% in 2010 to 22.6 K% came from. He also started relying less on his fastball (throwing it 52.6% of the time in 2011 compared to 57.3% in 2010) and a lot more on that dirty slider (32.4% in 2011 vs. 20.4% in 2010).
The point of all this? As pointed out, Bumgarner went under 20 K% in 2010 and over in 2011, and we can see this as the development of a tool that he is unlikely to lose. As long as his fastball velocity doesn’t drop out and he doesn’t forget how to throw the slider, he should be in line to post a K% close to that of 2011. One last reason – Bumgarner posted that strikeout rate in 2011 without a really workable changeup or curveball, posting a -5.4 wCB and -2.3 wCH in 2011, down from 2.4 and 5.0, respectably, in 2010. If Bumgarner rediscovers a decent curveball and changeup to keep hitters guessing, he could suffer a slight regression in slider value and still post a K% closer to 2011 than 2010. As a young pitcher on the rise, all signs point to Bumgarner posting a K% higher than 20%
Agree completely. The bump in K rate came from legit skill acquisition and there is no reason why he would lose that unless he made a conscious choice to tinker with the frequency of his slider. From a fantasy perspective the one thing that has to give with Bumgarner is his HR%. His park and division are great for low HR% but with his blah fastball I think he could see an uptick in HR allowed, which would have some impact on his ottoneu value in particular.
This guy is not a Bum. K rate(8.40) and walk rate(2.02). Those are Halladay type #’s. 6th in NL (4.15)K/BB rate. Great division, great park, 23 quality starts(Lincecum type #’s). I am taking the over as he looks healthy and might strike out 200 this year.
With everyone going over, I’m picking under…but just slightly. Even though last year wasn’t his rookie season, it was his first full season in the majors. I’m predicting somewhat of a sophomore slump (though it’s really junior). His K% will end up being around 19.5%.
As a totally non-statistical hunch, I don’t think I will pay what it will take to get Bumgarner in yearly leagues this year. For every SP who improves on his numbers every year age 20-25, there are 10 who experience big ups and downs. This kid is legit but my hunch is that his 2012 will not be as good as his 2011. I don’t foresee a sophomore “slump” necessarily, but I would bet against him improving on his 2011 numbers this year.
I’ll take the under – slightly. Bruce Bochy hates sabermetrics and he hates young players even more. Now that he knows there is a competition on the “interwebs” with the option of taking the sabermetric “over” on a young Giants players, he will make it his personal mission to ensure the “under”. He will order Bumgarner to throw a ball on every first pitch to a batter in low or medium leverage situation thus putting him in a hole. Bumgarner can still get close to a 20% strikeout rate with that 1-0 hole (60 K’s out of 319 TBF – for a 18.8% K rate in 2011) – but not quite. This means Bochy can win v the sabermetricians and still let Bumgarner pitch in high leverage situations.
I was prepared to take the contrarian “under” reminding everyone that pitching prospects don’t improve linearly year-over-year and that he was never a big strikeout guy in the minors (rates closer to 17%). But then I did a little digging, it became apparant that Over 20% is the right play.
First, let’s be clear about what taking the over means. In 2010, he had an 18% K-rate. In 2011, he had a 22% K-rate. Mbum can regress some from last year–call it “sophomore (or junior) slump” or whatever–and my bet will still be good. How much can he regress? Let’s change this rate stat to a counting stat. Last year he had 191 Ks in roughly 205 innings. If he faces the same number of batters, a 20% K rate means he would only need to strike out 169 batters (and, if we’re willing to round 19.5% up…165 batters) in that same 205 innings. Put back into rate stats, that means he’d need a 7.4 K/9 rate (which is Bill James projection). Is that doable?
Absolutely. Why?
1. He’s stopped throwing his Changeup. Excellent change-up hurlers like Hamels and Hoffman have roughly a 10 MPH difference between the FBv and CHv, as well as some deception in their delivery. In 2010, when MBum was throwing the Changeup 11 % of the time, the difference in velocity was 7 MPH. Although the pitch had a positive value in 2010, it actually had a negative value is 2011, which correlates with expectation. But the buried story here is that he’s replaced it (and then some) with a slider to righties which he’s throwing 3 MPH harder in 2011 than in 2010 and isn’t afraid to challenge them inside with it (per heat maps). From pitch values, it’s a much, much more effective pitch for him against righties than the change.
2. It’s warranted to be concerned about the innings jump from 2010 to 2011 and the potential FBv decrease. But here’s the thing. It was warranted to be worried about the jump last year too. He was even cited as a possible “Verducci effect” guy for the 2011 season. Fwiw, yes, there were definitely starts last year where he looked awful and tired and had a FBv under 90. He was hit hard. But you know what? He bounced back and righted the trendline over the last 6-8 weeks.
3. Is the O-Swing% sustainable? 30% is high, but it’s the “good” kind of high. Compare him to Gio Gonzalez as an example ofsomeone who’s O-Swing% rates might be more attributable to his wildness. MBum, at least by looking at his walk rate, appears to have much more control. A control pitcher with high O-swing% rates seems much more “real” or sustainable to me than a wild pitcher generating similar rates.
I’ll take the Over.
For funsies, I’ll take the under too (is this allowed?). If not, disregard.
Let’s start with a few admissions. The hype (Sickle’s “A” rating, BaseballAmerica’s love, etc etc) surrounding Bumgarner is very much deserved. He was drafted high, had an outstanding W/L records in the minors, and advanced quickly to the majors, where he more than held his own. But outside of the W/L minor league record and pedigree, his final 34 starts in AA and AAA don’t paint the picture of a pitcher primed for dominance. He had a 5.8 and 6.42 K/9 in his last two minor league stops and outperformed his FIP as well. His lack of “dominating” stuff in the minors be a red flag that his current K-rate might not be entirely sustainable?
So, people rightfully point to the development of the slider as the reason he’s K-ing more hitters. I think we can agree that without the slider, Bumgarner isn’t anything more than an average–maybe slightly below average–major league starter. Isn’t it possible that we–because of the hype that surrounded him–are prone to a sort of bias where our expectations are confirmed and thus we’re undervaluing the potential of the slider regressing some? This is, after all, not only one of the 3 best sliders in the league but a pitch that he barely threw until 2010! To me, that screams regression potential–where the NL, armed with a season’s worth of tape and an offseason to prepare might make it a less effective pitch. Bumgarner does, after all, generate a very high O-Swing%. What if teams start instructing players to be more judicious in their approach when they ID his slider, especially if he throws it early in the count or when he’s behind. O-Swing% will fall and so will K%.
Two of the comps that were made to MBum’s slider were Kershaw’s and Ervin Santana’s. This is interesting. Santana regularly has K-rates around 7 per 9IP and yet has a 2 MPH higher fastball average(although with less movement than Bumgarner’s, per PitchF/X). So, even if you believe his slider has a “true talent” of exceptional…having a high value slider alone does not guarantee Ks. If you look to Kershaw, his FBv actually dropped from Year 1 to Year 2 and then again from Year 2 to Year 3. While Mbum has gained FBv year-over-year, we regularly see pitchers–even elite 22 year old ones–lose velocity on their FB. If his FB goes back to being a minus value pitch (because of a decline in velocity), this probably doubly hurts him as he won’t be able to get into counts where he can use the slider effectively (because hitters will tee off on the FB early)…leading to a lower K-rate.
One other aspect of his underlying numbers that should stand out is his BABIP. It’s curious to see a guy of his ilk with a 20% LD percent, which does sort of support the .320 BABIP projections. While it’s tempting to want to regress this number, let’s remember that the OF defense for the Giants just downgraded to Melky Cabrera (-6.7 UZR in 2011) and Angel Pagan (-14.3 UZR in 2011). The Giants in 2011 were no worse than an average defensive team. It’s going to be more difficult making that claim this year, especially in the OF, and I can certainly see a scenario where more hits fall in and thus more batters are faced (by definition lowering the K-rate).
Finally, maybe his competition improves. It’s always been a chicken and the egg argument with the NL West. Is it the great pitching that makes the hitting look bad or is it the other way? The NL West’s hitting should be better than it was last year. The Dbacks have been trying to wring Ks out of their lineup since Kevin Towers arrived. San Diego has added intriguing, (mostly) young hitting and should score more than 593 runs this year. If Colorado is healthy, they’ll hit. Free from owner dysfunction, LA may look acquire a premium hitting talent (David Wright?) at the deadline. I’m not saying the division’s hitting is going to all of a sudden resemble the AL East, but maybe it’ll look a little less like… the NL West?
There are a lot of reasons to be bullish on Bumgarner going forward–especially longterm–but there are also reasons to think that his K-rate for 2012 might decrease enough to be below 20%.
let me shorten this post:
1. Slider value regresses
2. FB loses velocity, hurts OSwing%
3. Terrible OF defense lets fall more hits
4. NL West isn’t worst. division. ever.
Well done dude. Let me know if you don’t have access soon.
I’ll be taking the over on Bumgarner.
First off, look at his plate discipline. Bumgarner’s O-Swing rose by 1.5% in 2011, while his O-Contact fell by 2.4%. This is an indication that more batters are swinging at pitches outside the zone and whiffing on them. In addition, Bumgarners F-Strike rose by 2.4%. This increase is one factor that caused Bumgarner’s improvements in O-Swing and O-Contact.
Another reason that Bumgarner is going to break out next year and post a high K% is found in Justin Verlander. The same Justin Verlander who won the AL MVP this year. If you compare Verlander’s stats in 2010 to Bumgarner’s in 2011, you will notice a strong correlation.
Here are some of those prominent stats:
Verlander:
62.9 F-Strike%
9.3 SwStr%
31.9 O-Swing%
79.8 Contact%
8.79 K/9
2.85 BB/9
.286 BABIP
72% LOB%
Bumgarner:
62.6 F-Strike%
9.2 SwStr%
30.6 O-Swing%
80.3 Contact%
8.4 K/9%
2.02 BB/9%
.322 BABIP
72.4% LOB%
Look how similar their statistics are. Even though Verlander’s K/9 is 3.9% higher than Bumgarner, he makes up for that in BB/9 where he posts .83% fewer walks. The even more amazing part of this is that Bumgarner did this in his second year of the league with a .332 BABIP compared to Verlander’s .286. If Bumgarner follows the natural path of regression to the mean, then he will be luckier and improve simply because of that.
The major argument against Bumgarner getting above 20% K% is that he is in a particuraly good division. However, the talent within the division helps Bumgarner’s case. In 2011, the other 4 teams in the NL West had an average of 80 wins. The AL Central in 2010, Verlander’s division, had an average of 78.25 wins which means that Bumgarner posted statistics equal to Verlander in 2010 in a harder division.
Madison Bumgarner also happens to be on the Giants, a team that scored fewer runs than any other team last year besides the Mariners. However, Bumgarner is likely to receive more run support this year with the addition of Melky Cabrera, the health of Buster Posey, and the additional playing time for Brandon Belt. Others may say that the Giants aren’t very adept on defense, but we are talking about K%, a stat that is fielding independant. If you go by fielding independant statistics like FIP, you will notice that Bumgarner had the fourth-lowest FIP at 2.67. That puts him ahead of teammates Tim Lincecum and Matt Cain, along with Justin Verlander.
The last reason why Bumgarner’s K% will stay above 20% is found in PitchFx. There you can see that the velocity of all his pitches he throws more than 1% has gone up. Bumgarner has also decreased the frequency he throws fastballs, resulting in a very diverse spread of pitches with no one pitch being relied upon. This is also the same in Roy Halladay, who like Bumgarner, throws most of his pitches between 30% of the time and 10%. Also, if you compare Halladay’s and Bumgarner’s heat maps, then you can see that when comparing both vertical and horizontal movement of pitches, more of Bumgarner’s pitches are in quadrent 1 than Halladay, whose pitches are spread out over quadrent’s 2 and 4. I’m not saying that Bumgarner is a better pitcher than Roy Halladay. I’m just pointing out that there are many similarities between the two pitchers.
From all of this, you can see that Madison Bumgarner has no where to go but up… except in ERA where he will join his fellow Giants pitchers. The even more amazing part is that there is so much more I could have written about him. There is a plethera of information on Bumgarner out there and almost all of it supports him.
If you made it this far in my comment then I applaud and thank you for reading this ranting of a sabermetrics addict.
When looking at the elite strikeout pitchers, I like to see a low Contact%, high SwStr% and a dash of decent velocity potential. Looking at Bumgarner, he does well in all three areas even if he may not be particularly outstanding in any of them.
With the Contact% wavering around 80%, it separates him from some strikeout aces who might be able to come in at 75% here but it’s still quite good. Of pitchers with over 100 IP and a K% above 20%, there were 12 that had a Contact% at or above 80% so it’s not outlandish for him to keep that rate and be above a 20% K-rate.
The SwStr% is also good but not phenomenal at 9.2%. The pitchers that are higher up on the food chain here might be well above 10% but that’s certainly not a prerequisite to achieve a mark of a 20% K-rate. There are quite a few pitchers with a 20 K% or more clocking in with SwStr% between 8-10% . For the final fact, his fastball velocity of 91.7 MPH compares well without other strikeout artists and isn’t a cause for concern so he’s got that going for him too.
If the over/under was his 22.6% rate from last year then I would take the under as there’s seemingly good reason that he’s not among the most elite strikeout pitchers out there. But, 20% is definitely reachable for someone with his pedigree and probably about where he should be each year.
I have a quick easy analysis I use…I figure Bum for over 200 innings averaging 5 batters per inning…if he gets 200 k’s that will put him under 20%.
that would be a WHIP over 1.40 !!! If that happened he’s got bigger things to worry about than his K-rate.
Hey everyone! Thanks for playing. I’m going to give jcxy extra credit for arguing both sides, but this is a one-time deal. Let’s not all start arguing both sides, that’ll get confusing.
Congrats jcxy, you should find that you can access FG+ shortly.
We’ll do this again every day this week, so everyone else will have another shot! Good luck.
thank you rotographs staff!!
i can’t wait to access fangraphs+!!!