2017 Pod Projections: Keon Broxton

The Pod Projections are back! My projections are based on the methodology shared in my eBook Projecting X 2.0, and the process continues to evolve and improve.

2017 Pod Projections Index:
Lance McCullers
David Dahl

Who would make for the perfect hitter to be Pod Projected? The one with the biggest difference between the Fans and Depth Chart projections, of course!

Biggest diff between fans & depth charts:
1) Broxton
2) CSeager
3) TimA
4) Drury
5) CHernandez
6) Yoan
7) Eaton
8) GaryS
9) Bryant
10) Votto

— Eno Sarris (@enosarris) February 28, 2017

The Fans projections are notoriously bullish, but sometimes they rightly believe in a breakout, whereas the projection systems are programmed to forecast severe regression. Keon Broxton is no Spring chicken and is already 26 heading into the 2017 season. But he got his first chance to play regularly last season and made the most of it by posting a .343 wOBA, displaying both power and speed, excellent plate patience, and playing fabulous defense. Naturally, everyone is skeptical, though the Fans are far less so than Steamer and ZiPS. What about the Pod Projections, you ask? Let’s find out!

Plate Appearances: 527

My projection assumed that Broxton would open the season batting sixth. The thought was that Domingo Santana would slide into the two hole, with Ryan Braun, Eric Thames, and Travis Shaw hitting third through fifth. But now there’s managerial chatter suggesting Broxton may wind up beginning the year at the top of the order.

Counsell on Broxton’s spot in lineup: “If we can get Keon going, we'd love to hit him in one of those spots at the top of the lineup."

— Tom (@Haudricourt) February 28, 2017

In all four games he has started, Broxton has batted second twice and first twice. If it does become clear that he’ll hit atop the lineup, it would be huge for his fantasy value and all his counting stats would get a meaningful boost. It will take another week or two of games to convince me the top of the lineup is his most likely spot before I adjust my projections.

BB%: 11.1%

Broxton posted double digit walk rates at times during his minor league career, though never the 14.8% mark he posted with the Brewers. His strong walk rate was driven by a below average Swing% and impressively low O-Swing%. Also aiding his walk rate was a relatively high SwStk%. He wasn’t swinging often, and when he did swing, he didn’t make great contact, which would extend at-bats, and lead to more walks. Given his plate discipline metrics, I would expect him to continue walking at a well above average clip.

K%: 31.4%

So the one huge flaw in Broxton’s game last season was that ugly 36.1% strikeout rate. He also struck out often in the minors, generally in the 25% to 30% range, so we can’t even point there to provide us with hope for improvement. Still, his 14.7% SwStk%, while poor, wasn’t atrocious, and it certainly doesn’t match with a mid-30% strikeout rate. In fact, his xK% (a metric I developed years ago, but have since forgotten about and rarely use nowadays), was a slightly more palatable 31.3%. It’s just a coincidence that my projection is nearly identical to his 2016 xK% mark as I didn’t even calculate it until this very moment! Anyway, like most young hitters, he’ll probably improve a bit on his contact rate and I would think he starts swinging more at pitches inside the strike zone.

GB%/LD%/FB%: 46% / 23% / 31%

Last year, he posted a 25% line drive rate, so I simply forecasted a bit of regression there and added most of the loss to his ground ball rate.

BABIP: .345

Here’s one of the primary sources of disagreement between projections. Broxton posted a .373 BABIP last year, which would have ranked him fifth in baseball if he qualified for the leaderboard (and maintained that BABIP over all those additional at-bats, of course). Naturally, a rookie with an inflated BABIP is going to be expected to regress heavily the following season. But what if I told you the following two facts:

1) His minor league BABIP over the last three seasons has averaged .369, including two stops over .390.
2) His xBABIP was a robust .381. Yes, that’s higher than his actual BABIP. And you thought a .373 was ridiculous? xBABIP thinks he actually experienced a bit of bad luck!

That xBABIP was the second highest 2016 mark on my spreadsheet, behind just Adam Frazier, who tallied just 146 at-bats. It was fueled by a high LD%, pop-up avoidance (just one over 207 AB), an elite Hard%, excellent speed, and the lack of any pulled grounders into the shift. Essentially, he was good at everything xBABIP factors in. So while BABIP regression simply must be assumed given the small sample we’re working with, everything points to him being another Starling Marte type, with obvious BABIP skills.

HR/FB Ratio: 15.0%

Power was never a huge part of Broxton’s game, but was always in his bat, as he hit 19 homers in 2012, and generally posted ISO marks in the .150 to .200 range. But no one saw a 25.7% HR/FB rate coming! We all pretty much know that’s going to collapse, but how far does it tumble? That’s what we have xHR/FB rate for! His 11.1% Brls/BBE was fantastic, but he pulled the ball and went the opposite way at a below average clip. That said, his xHR/FB rate was 17.1%, which does signify the risk of major decline, but still supports the notion that there’s real power here. Playing home games in the third best park for right-handed home runs is just a bonus.

Runs and RBI: 59 and 49

For whatever reason, Broxton scored runs at a poor clip last year, which is especially head-scratching given his speed. I am projecting improvement in his runs scored and runs batted in rates, but again, these forecasts assume he’ll be hitting sixth, and look highly unimpressive from a fantasy perspective. If he does move to the top of the order, expect a significant increase in runs scored and a more modest increase in runs batted in.

SB: 33

Broxton attempted a steal once every nine plate appearances (for context, Billy Hamilton attempted one every 7.9 plate appearances and teammate Jonathan Villar tried one every 11 plate appearances). That’s a pretty high rate and it would be silly to expect a repeat. My projection assumes a steal attempt every 12.3 plate appearances, which is much closer to his recent minor league marks. He was also successful on 85% of his attempts, so he should continue to run freely.

Below is my final projected fantasy batting line, along with the other systems for comparison:

After talking up his power and utilizing his xHR/FB rate, turns out I’m actually the lowest on his projected ISO and home run total! That’s quite a surprise.

The real differences come from two places — strikeout rate and BABIP. Bizarrely, ZiPS is projecting a strikeout way above everyone else and even worse than Broxton’s 2016 mark. I’d love to learn what inputs drove that forecast. That’s predictably killing its batting average projection.

Then we have the BABIP, a metric the Fans and I are far higher on than the computer systems. I’m leaning on xBABIP to inform my projection, and perhaps the Fans have all read about the components of the equation and agree with its conclusions? Broxton will make for a good test of the predictive value of the new xBABIP, even though it’s supposed to be used in a backwards-looking fashion.

Even with a .345 projected BABIP, it’s clear that there’s serious batting average risk. However, there’s intriguing power/speed skills and the real possibility that his counting stats enjoy a bump if he moves toward the top of the order. He’s being plucked as the 39th outfielder off the board and 182nd overall in NFBC drafts, which seems like a reasonable price. If you’re in an OBP league, he’ll almost surely be a bargain.