Five questions: Cincinnati Reds

How much better will the fielding be?

Last season, the Reds had the worst fielding in the National League. They were last in the league in DER (0.674), last in the league in THT’s plus/minus team fielding plays (-53 plays), and were last in the league in bUZR (-45 runs).

This season looks to be better. Two of the worst defensive players on the team departed last season in Ken Griffey Jr. (-11 runs with the Reds, according to bUZR) and Adam Dunn (-18 runs). And assuming Alex Gonzalez is healthy and competent, Jeff Keppinger (-15 runs) looks to get far less time at shortstop this season, where he has been playing out of position since 2007. Those three players alone, even in part time play, were the difference between an average-fielding team and a horrible fielding team. And all things being equal, replacing those performances with average fielding could net the Reds another four wins.

Finding average fielders to insert into a lineup is no easy task, however, if they’re going to hit at all. What can we reasonably expect from the Reds in 2009? We can get some idea of this using CHONE’s defensive projections and some best guesses on playing time (some details on what I did are found at the end of the article). Ultimately, I came up with an overall projected team defense of approximately nine runs below average. Maybe they won’t be a plus defensive team—at least not with Edwin Encarnacion and Jonny Gomes figuring to get a fair bit of playing time—but they can still be expected to post roughly a three-win improvement over last season.

How bad will the offense be?

Departing with the defensive liabilities of Dunn and Griffey was a substantial amount of offense. So are the defensive gains negated by the loss of offense?

No hitter on this team figures to be as productive as Dunn will be in 2009. Dunn’s an excellent, if often miscast hitter, and the Reds don’t have anyone who is likely to approach what he can provide with the bat this season. But the surprising thing is that even despite not having Dunn around, the Reds’ offense might actually be improved in 2009.

How? It’s addition by subtraction, mostly.

The Reds gave 20 percent of their plate appearances in 2008 to the trio of Paul Bako (.217/.299/.328), Corey Patterson (.205/.238/.344) and Keppinger (.266/.310/.346). All hit below “replacement level” (not accounting for position), to the tune of 25-30 runs. The first two are no longer with the team, and Keppinger should receive less playing time this year while also showing a bit of a rebound at the plate. If those performances are replaced on this year’s team with even replacement-level hitting, the Reds stand to gain two to three wins.

To get a more objective estimate of the 2009 Reds offense, I looked again to CHONE projections and my best guesses on playing time (again, see methods at end). This is admittedly very prone to error, especially among the principal bench players, as it’s very hard to predict how playing time will shake out over a 162-game season. But I think what I’ve done is pretty reasonable, and different scenarios among bit players usually make very little difference to teams’ offensive projections.

Overall, after working through the projections and forcing the plate appearances to match last years’ Reds, my estimate for 2009 Reds offense was…. 25 runs above average. If the NL average is 734 runs scored per team, as it was last year, that would put the Reds at 759 runs scored. That would have ranked sixth in the league in 2008. And this is from a team that scored just 704 runs last year with a partial season of Adam Dunn. Granted, if you adjust for park effects, you should cull 15 or so runs off of this total. But even so, most of us think of the Reds as a below-average hitting team these days. These data indicate that this isn’t the case.

It took me a while to buy into this. But I’ve repeated this several times using several different approaches, and my estimates are always within 20 or so runs of this total (and often are above the number I’m reporting), which has to be within our margin of error here. What’s driving it? Strong performances by Joey Votto, Jay Bruce, Encarnacion, Brandon Phillips, and new addition Gomes (I’m giving him 327 PAs as part of the left field squad, and his CHONE projection is very strong).

Perhaps most importantly, we get only a handful of sub-replacement level hitting performances. Yes, some of the these guys will miss low on their projections. But others might overperform, and CHONE’s been as good a projection system as there is, historically.

What about the rotation?

The Reds have quietly assembled what could potentially be one of the deepest rotations in the league. They have six starters with a good shot to be at least respectable, and at least one in Edinson Volquez who has a chance to be brilliant. The competition for the fifth spot is as tight as it gets between once uber-prospect Homer Bailey and the more experienced hitting curiosity Micah Owings, but regardless of what happens by Opening Day I expect both pitchers to get a good chunk of starts this season.

To project rotation performance, I took CHONE projections for the top four starters as-is: Volquez, Aaron Harang, Bronson Arroyo and Johnny Cueto. That took care of 116 games. The remaining games started were then filled by the next guys in line, and I allocated 20 starts to Owings, 20 to Bailey, and two each to Matt Maloney, Ramon Ramirez and Daryl Thompson.

Altogether, these starters account for 162 games started, 939 innings pitched, and a … 4.15 ERA. Last year, Reds starters accounted for 917 innings pitched and a 4.97 ERA, so this is a massive projected improvement. The worst CHONE projection among these pitchers is Bailey with a 4.61 ERA, but to be honest the projections for the six primary starters look pretty reasonable to my fan’s eye. (Owings is the only one that looks overly optimistic to me.)

Altogether, this group of starters is projected to allow 468 runs in 939 innings, which is much better than average.

What about the bullpen?

The bullpen also looks pretty solid, as it was last year. Francisco Cordero may be overpaid, but he’s a good anchor to a pen that also features a couple of good young arms in Bill Bray and Jared Burton, as well as “veteran presence” players David Weathers and Arthur Rhodes. Bray and Burton should be the better pitchers of those four, but will likely have to defer to the old guys until or unless they flame out. Even so, this is a good core of relievers, with other guys like Mike Lincoln and Nick Masset who should be decent middle relief contributors.

To continue the projection work from the prior sections, I sequentially added CHONE projections of relievers in order of my perception of the likeliness that they’ll get a shot at contributing this season. I needed 503 innings, and I had to add 10 relievers into the mix to get them, all but the last contributing between 30 and 70 innings. Altogether, CHONE projects those relievers to have a 3.99 ERA (last year, the Reds’ pen sported a sparkly 3.81 ERA), allowing 241 runs in 503 innings. Essentially, that’s holding steady on what was already a good bullpen last season.

That makes the overall pitching look pretty good: 1,442 innings, allowing 709 runs (~25-35 runs less than an average staff in the NL), which works out to ~4.4 runs per game. That’s shaving just shy of 100 runs off of last year’s total, and would have been good for fifth in the league last season. Most of that’s the starting pitching, but I’d wager that 20-30 runs of it is due to improved fielding. (Rally reports that these projections already take into account something about team fielding, so I won’t adjust further).

So just how good are the Reds going to be this year?

When Rally published his NL Central projections last month, he had the Reds one game over .500. The Hardball Times Preview’s projections had them one game under .500. The consensus is that the defense (pitching and fielding) looks very promising, while the offense is probably a bit below average.

With the estimates I came up with here of 759 runs scored and 709 runs allowed, and plugging them into the pythagoran formula, I’m getting an overall expected winning percentage of .534. That would make them 86-75, and surprise contenders in the NL Central.

My guess is that I’m still missing high for some reason, and probably on offense. Maybe I still need to include still more PAs by reserves or something. But the real point of all of this is that this new model of a Reds team—one supposedly built more on pitching and fielding than offense—might surprise some people this year. They’re younger, faster and better in the field than they have been in a long while. They have a good-looking young pitching staff with genuine talent. The offense might also turn out to be at least average in its run output, provided it can avoid black holes in the lineup. All of this means that the Reds are likely to be at least a .500 ballclub, and if things go their way they could find themselves in the hunt this September.

We haven’t seen that kind of thing in Cincinnati in a very long time, so here’s hoping.

Some additional details on methodology

To do playing time projections with position players, I took all starters and used every PA that CHONE projects. I then divided up other plate appearances among reserves. For catchers, I have Ryan Hanigan getting roughly twice as much playing time as Wilkin Castillo, but I’m expecting that Castillo will get some PAs. Infielder reserve PAs were split among Keppinger, Danny Richar, Adam Rosales and Paul Janish, with a double share going to Keppinger given his status.

Outfielders were handled a bit differently, given the chance for a three-way “platoon” among Chris Dickerson, Gomes and Jerry Hairston. I split the starting LF job evenly between them. I then split the remaining reserve PA’s between the three left fielders plus Norris Hopper. This gave Dickerson, Gomes and Hairston about 75 percent of their projected CHONE PAs, while Hopper got just 37 percent of his.

To estimate the impact of pitchers on offense, I used 2008 league-wide pitcher hitting performance and pro-rated it down to the 326 PAs I needed to complete the Reds’ offense after the position players were accounted for. No effort was made to account for the Micah Owings factor. Overall, not counting the generic “pitcher” I added, 17 players were given projected playing time, all of which received at least 90 PAs. Last year, 25 players served as non-pitchers for the Reds (including Owings), but only 16 got to the 90-PA mark.

Fielding was based on CHONE defensive projections, pro-rated to playing time. To deal with position inequities, I based each players’ fielding ratings on the position assigned in CHONE, but then used position adjustments to get them all on the same playing field. This way, for example, if Keppinger plays lots of shortstop (+7.5 runs/season bonus) instead of third base (+2.5 runs/season bonus) where he should be, his overall fielding rating will be only marginally affected even though he’ll compare badly to other shortstops.

Reds hitter run production was estimated using Tom Tango’s linear weights. I ran a check with base runs on overall team hitting statistics and didn’t find much of a difference between base runs and what the linear weights project, meaning that there aren’t extreme enough OBP vs. ISO issues to cause linear weights to be a poor predictor of team-level offense. Pitcher run allowed estimates are straight from CHONE, pro-rated by innings.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: What is position scarcity?
Next: Five questions: St. Louis Cardinals »

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>