Washington Nationals: Predicting The Statistics Of Starting Pitchers

2 of 6

Predicting the statistics of players is a time honored tradition. What do the computers think the Washington Nationals pitchers will do in 2017?

The hopes of success for the Washington Nationals rest on their starting rotation.

Worth hundreds of millions, the five men taking the hill to start the game set the pace and tempo for what is to follow. Their production sets the table for the offense to follow and what the bullpen can accomplish at the end of the game.

How will they do this year?

We know the pressure is immense to not simply advance in the playoffs but win the World Series. Aside from luck, which you cannot throw into numbers, we can see what the computers say might happen.

Fangraphs—that oasis of sabermetric goodness—has three different projections published on site for every player. From Dan Szymborski’s well-known ZiPS, to the Steamer system created by students from Brooklyn a few years ago mixed, finally, with FanGraph’s own mix.

From these three projections, we can get a look at how any player might perform before the year. Does it mean anything? No. But, it offers a guide for what might happen. Although the number-crunching is accurate enough to be taken seriously, it cannot predict injuries or hot and cold runs.

With that disclaimer, the site still predicts Washington to win 91 games, the National League East by five over the New York Mets and earn the third-best record in the league. Sorry, this time the Nats travel to Los Angeles for the start of the NL Divisional Series opener.

Here is what the computers say the projected starting five will do for numbers this year. Come October, we will either laugh or cry.

MAX SCHERZER

W/L: 15-8 ERA Range: 2.99-3.18 IP Range: 203-210.7 WAR Estimate: 5.6-5.8

All three projections have Max Scherzer starting three less games this year than last at 31. An 18-game winner, a 15-win effort is a smart conservative effort. The ERA projection is right along last year’s 2.96 while his K/9 numbers drop from 11.2 to 10.8.

One thing all the models say is concerning is a raise in Batting Average for balls in play. Last season, Scherzer’s BAbip was a healthy .255. All three models show a jump to around .300. Steamer gives the best estimate at .289 while Depth Charts has him at .304.

If true, that puts added pressure on Scherzer to keep the ball in the yard, a problem he has. They all have him dropping to around 1.0 HR/9. A good sign.

What the computers cannot project with Scherzer is his ring finger injury. Even with three less protected starts, he has yet to throw an inning this spring and is avoiding using his normal grip until the pain subsides.

Although the fastball speed appears to be no different in side sessions, the only true judge is game action. If he feels comfortable using the three-fingered approach, Scherzer can take the pressure off his fractured knuckle and relax.

The WAR estimate puts him in contention for another Cy Young. If the Nats are to repeat in the division, these numbers need to be accurate. They need Scherzer to pitch into the seventh every time out.

TANNER ROARK

W/L: 11(12)-10 ERA Range: 3.85-4.25 IP Range: 170.7-177 WAR Estimate: 1.8-2.0

The models do not love Tanner Roark.

Because these projections are weighted over the last few years, and not on last season entirely, they look at his 2015 where he pitched mainly out of the pen. Since he posted a 4.38 ERA that season, the models have a hard time compensating between his weaker year and last year.

If he regresses back to 2015, the Nats are in serious trouble.

What the projections agree on is he makes 28 or 29 starts. That is a number a bit low. Their call for him to pitch under six innings a start is puzzling too. In 2016, Roark averaged nearly seven a start.

Only Steamer thinks he matches his 7.4 K/9 rate at 7.2. ZiPS and Depth Charts have Roark under 7.0. ZiPS puts him at 6.6. All three projections list his BAbip at a minimum of .290 with Steamer calling for a .305 rate. Last year, the number was.269.

The computers say Roark is headed for a major down year, but with his career bouncing roles from starter to reliever, it is hard to shape out an accurate call. If he performs at these levels, it is bad news.

Here, the projections feel wrong. Roark should make at least 30 starts and pitch 200 innings.

STEPHEN STRASBURG

W/L: 13-8 ERA Range: 3.11-3.20 IP Range: 159.3-177 WAR Estimate: 4.2-4.4

All three models are bullish in Stephen Strasburg’s return.

ZiPS has him at 27 starts while the other two give him 29. Strasburg has not started over 27 since his 34 in 2014. Although all three models suggest he pitches under six innings a start, again he has not averaged over six since 2014.

Where the projections are strongest is with ERA. Two injury-plagued seasons pushed his real-life numbers to 3.46 and 3.60 since 2015. The models have Strasburg back into peak range slightly over three.

To get there, Strasburg must fan his fair share of batters. The models agree, setting his K/9 rate around 10.6. Last year, before the arm trouble kicked it, it was 11.2. Strike outs are his bread and butter.

BAbip is not. When not getting batters to whiff, opponents hit Strasburg hard. His .294 rate last year was his best since 2013. The models have him in his 2014-15 mode, projecting anywhere from .311-315. A precise number.

If he stays healthy, Strasburg will beat his innings estimates and his start projections. Because he pitches well despite a high BAbip rate, he could exceed his WAR estimate of 4.3.

What we want to see is someone who takes the mound every fifth day who can stop a losing streak. Strasburg’s switch to pitching full-time out of the stretch is interesting. If it simplifies everything, could he take stress off his arm and pitch to contact more?

GIO GONZALEZ

W/L: 11-9 ERA Range: 3.75-3.86 IP Range: 163.3-165 WAR Estimate: 2.9-3.1

The computers love Gio Gonzalez. In unison, they look at last year’s 4.57 ERA and reject it as an anomaly. All three are in lock step with his numbers. They see him as the Gonzalez from 2015, someone who pitched overall better than his record.

If early spring returns are a measuring stick, the projections are solid. He looks the confident pitcher again.

As with the other predictions, innings pitched projects conservative. Yet, they are within a tenth of each other on ratio calls. With his K/9 numbers, Steamer has him at 8.35, ZiPS at 8.60 and Depth Charts at 8.5. Last year’s true number was 8.7.

Although they think he improves his ERA by over a half-run, they show his BB/9 rate climbing from 3.0 to 3.2 and his BAbip near 2016’s .316.

If you were offered the projections above, you would take them and run. As a No. 4 starter, anything under a 4.00 ERA is super. Again, they project him missing three starts. If Gonzalez makes over 30, the inning total will creep toward 180.

Six innings every five days while giving the Nats a chance to win is what you want from him. There will be nights where the fingernails are gone after Gonzalez is done, but when we look back, those numbers are reasonable targets and okay.

JOE ROSS

W/L: 9(10)-7(8) ERA Range: 3.59-3.79 IP Range: 133-148 WAR Estimate: 2.2-2.5

Because of his experience and injuries, all three projections cannot get a grasp of what Joe Ross should do.

More from District on Deck

Because he has so much movement on his pitches, Ross is never a guy that will shut down the side quickly. He will hit 100 pitches a start fast, infuriating fans as he struggles to get into the sixth on his best nights. He relies on that movement for outs and gets burned.

Still, a healthy Ross rill pitch 160-180 innings this year. The record and ERA are around his capabilities but, in 26 projected starts, he should do better than a flat five innings per.

All three models see improvement in BAbip and K/9. From a 7.97 last year, Steamer says 7.9 while the others have him over eight. ZiPS the most bullish at 8.12. As he strikes out more, hitters will fare worse making contact.

From a .319 mark last year, Depth Charts has him at .312, ZiPS at .309 and Steamer an impressive .303. If Ross improves there it translates to an extra couple outs per start. As the fifth starter, how well he does lets the bullpen rest longer.

If, and that is big in Ross’ case, he can get 18 outs a start, the pen will need three innings to finish. For a back end starter that is tremendous. As with Gonzalez, the longer they keep the Nationals in games and pitch, the less stress and fatigue the bullpen has.

Next: Previewing Jayson Werth's 2017

Come October, a few innings here and there makes a difference for games inevitably won or lost not be starters, but tired bullpen arms.

Home / Nationals News