Dan Szymborski over at Baseball Think Factory annually posts his ZiPS baseball player projections. It appears that he now has projections posted for 29 of the major league teams, including all of the National League Central teams. With lots of recent discussion here about how the Reds will fare against the Cardinals, Brewers, and Cubs this season, I decided to compile the projections for each of the NL Central teams and use them to compute the final standings for the upcoming season.

The methodology I used was to compile a full seasons worth of AB (about 5500) for each team ,comprised of their projected starters and reserves. I then computed Runs Created (RC) from the compiled team stats to represent the number of projected runs scored for the offense.

For pitching, I compiled pitching stats for the projected starters (about 960 IP) and the projected relievers (about 480 IP). I then reverse engineered Runs Allowed (RA) from the computed ERA for the entire pitching staff.

Finally, I used Bill James Pythagorean expectation formula, plugging in the RC and RA numbers to compute the expected number of wins for each team. Here are the results:

Team Pyth. WP W L RC RA OPS ERA
St. Louis 59% 95 67 800 667 .767 3.79
Milwaukee 56% 91 71 774 681 .758 3.88
Cincinnati 55% 89 73 772 697 .756 3.96
Chicago 53% 86 76 765 721 .753 4.09
Pittsburgh 49% 80 82 773 784 .756 4.83
Houston 45% 72 90 680 756 .709 4.30

The ZiPS projections view the Cardinals as a tick above the Brewers, Reds, and Cubs, who are all packed tightly near the top. The Reds, who are projected here at just two wins off of last season’s pace, would not even make the playoffs as a wild card with the third place finish. Milwaukee’s pitching acquisitions over the winter have definitely, on paper, propelled them into the mix for playoff contention. This also reaffirms that the Cubs do still have a very solid team, and could be in the mix if a few things broke right for them for a change.

These projections, while not quite what Reds fans were hoping for, do show that we are in for another exciting season in the NL Central.

*Related*

In my mind, I think the fact that this projection has the Pirates as only two games below .500 instantly destroys any credibility it had to begin with.

That worries me too, Tom. I’d have to see the rest of the NL, obviously, but I have doubts about any projection that puts the NL Central at a combined 54 games over .500. Last year they were 48 games UNDER .500.

Since we know the division will play .500 ball against itself, and the division schedule is roughly 79 of the 162 games, that means that the NLC teams would have to play the rest of baseball at 276-222 (46-37 each, on average). That’s against the Phillies, Giants, Braves… and the AL East as interleague opponents. Unless the Nationala come in at 0-162, it’s hard to make the math work.

Also, did you use the RC formula that incorporates SB?

Since it’s a simple Pythagorean expectation projection for the records, it’s not going to add up. But on the other hand, it’ll be no big surprise to see 4 of the teams in this division are above .500, and while Pittsburgh is not good, they aren’t going to be 57 wins bad again either. I just used the basic RC formula.

I wouldn’t get too hung up on the records. For me the exercise was more to see the general potential strengths of the teams in the division. I think it succeeded in showing that it’s going to be a four horse race in the NL Central next season.

80 wins for Pittsburgh?

@WishboneD: I should’ve read the comments first…

While I don’t think Houston is going to be great, they did play over .500 baseball from June on last year after that faceplant start. I think their pitching could be decent to good. For them to be better though the young players are going to have to hit.

@Chris Garber:

Don’t get too hung up on the actual records they’re projecting. These projection systems are pretty good all things considered, but they can’t predict how runs are going to be distributed game to game (i.e. you lose 2-1 one game, win 10-3 the next, then lose 1-0) so they’re really just rough estimates of how many games you’ll win. Look more at how close teams are to each other.

How did you allocate playing time among the player projections? Is this a comparison of starting lineups, and 5 primary starter/6 primary relievers? It seems the RC per team is based more on a starting 8 type of projection…

You’ve touched on two of the things I’ve thought of that would play into changing the shape of these numbers a bit since I put them together. I’m sure there are lots more, as Chris G. has already pointed out that using the more advanced RC formula using SB would have been better. 🙂

The playing time allocation was factored in. The projected starters and bench players AB are weighted appropriately within the 5500 AB for each team. What I did forget, however, was to allocate about 300 AB of a .350 OPS for each team’s pitching staff. So I’d need to substitute that for 300 AB worth of bench players who would double that production for those 300 AB. That would lower the RC for each team and reduce the expected wins for all teams. Using PA or a close approximation of AB+BB, instead of AB, would have probably been better too.

Likewise, I attempted to identify the projected starters and relievers for team and weight a reasonable number of innings for each player to accumulate a total of 1440 IP total, 960 IP for starters and 480 IP for relievers. So bench and bullpen depth is reflected in these numbers.

Also, when I reverse engineered RA from the pitchers ERA, I did not take defense into account, I simply assumed a flat correlation of .91 between ER and R for each team, which was last years NL average. So basically each teams RA was computed with an NL average defense. The ratio for the NL Central teams actually ranged from the Cubs at .87 to the Reds at .94

Another can of worms would be in how the projections produce the ERA totals themselves. Do they take defense into account when determining them? Not knowing, it may perhaps be better to base the RA numbers off of XERA (expected ERA) which computes the players expected ERA based on his peripheral numbers to remove defense from the equation all together.

There is lots of room for improvement on these and we could pick nits on them forever and a day, but like Brien mentioned above, the projection systems do a decent job all told, so mostly just pay attention to the on paper groupings to see who we need to keep any eye on as the season progresses.

What did ZIPS project last year?

I think a lot of people are overestimating the Cardinals this year. The fact that they faded so steeply last August and September, to me, says they are trending downward. I still believe McGwire is a negative factor—-not just his coaching philosophy, but his very presence is begging for negative karma. You can’t swim around in sewage and still smell like Old Spice.

The Cardinals starting pitchers are also getting older. Their bullpen is really just kind of average. They don’t have much of a bench this year. And no matter what they expect, Pujols’ contract situation will be a distraction

It will be interesting to see how much of a drop-off the Cardinals have defensively from losing Ryan at SS and playing Berkman in the outfield.

I agree with you. Given the cards rep on pitching down in the zone and generating groundballs, the Ryan dump seems very odd with his plus-glove at SS. Picking up Theriot to platoon with Schumaker made some sense and provide that Larussa-loving double-switch flexibility but instead its to replace Ryan. and then to think berkman in RF will survive defensively. Cards fans last year were moanng and groaning about Randy Winn in RF, I can only imagine what they will be saying about Berkman.

oh well, would seem to be good for the Reds. I can only imagine what Carpenter is going to do with Theriot after watching groundballs go by he was used to seeing Ryan scoop up.

Go 2011 Reds!

@doctor: More to your point, MLB Network was doing a profile on the Cards and said they finished 1st in ERA last year but 13th in the NL in strikeouts. So yeah, they need to field a team that can catch the ball. Problem for them is that they went out and acquired players who can’t do that. The Cubs are terrible defensively and didn’t improve either. The Reds by comparison could potentially be even better because Hanigan, Janish, and Lewis will get more playing time than the below average defenders they replace. Advantage Reds.

@doctor:

Somehow, I don’t see Theriot putting up with any of Carpenter’s bullying shit the way Ryan was forced to. If Theriot needs his glove replaced at the last minute, he’s not going to put up with Carpenter giving him a public dressing-down because of it

But now you all mention them adding Berkman (and Theriot, for that matter), the Cardinals are also trending older. The Cardinals starting line up has at least 3 players over 30. Except for Hannigan and Rolen, all the Reds starters are under 30 and one is under 25

@Python Curtus:

Right. Any time you have a mass murdering, kitten-killing, taking-candy-from-a-babyin’ person like McGwire around, you’re just begging for cosmic retribution.

This projection looks pretty bad to me, to be honest. Not knocking the effort, I think you just had bad input projections, but how can you justify projecting the Cardinals having the best OPS in the division?

The Reds had the best OPS in the NL last year at .774, and the Brewers, who have a good offense, were third at .754. The Cardinals were 7th in the NL at .733. Now somehow they’ve moved past both teams and added .034 points. That’s huge leap for a team.

This jump comes from what, an ancient Berkman? Ryan Theriot? I just don’t see it. I would project the Reds and Brewer’s offenses to be pretty close, and the Cards to be a step behind, same as last year, since none of the three made significant changes to their offenses.

The Brewers will probably be in the mix more this year because they improved their pitching.

This seems like a time when the stats get in the way of common sense.

Here’s the ZiPS projection OPS numbers for the Cardinals and the Reds starting eight for your viewing pleasure.

Cardinals OPS

Molina 0.706

Pujols 1.011

Schumaker 0.698

Freese 0.727

Theriot 0.651

Holliday 0.867

Rasmus 0.784

Berkman 0.828

Reds OPS

Hanigan 0.720

Votto 0.936

Phillips 0.759

Rolen 0.785

Janish 0.660

Gomes 0.773

Stubbs 0.717

Bruce 0.835

that predicts a 100 point drop for votto. i know it’s guessing regression to the mean, but that doesn’t really make sense for a player who is entering his prime. the mean includes his first years, which are usually bad for everyone relative to their prime years. if anything i’d guess a regression to his mean of the last two years, which would be about 1.000.

stubbs ops’d .773 last year, i see no reason to project a massive decline this year. if anything, he’ll probably be more like he was in the second half, which was .857.

that’s also a decline for bruce, which i think conventional wisdom disagrees with, but it’s at least close to last year.

that projects a 50 point increase for berkman, and i’ll believe that when i see it. he looked like a man in decline to me last year, and playing all of his games in the outfield won’t help.

i disagree with those projections, and stand by what i said before: beyond ginned up guesses from a stats robot, i see no compelling reason to think the cardinals will add 35 points to their team ops, or why their offense would pass the offenses of both the reds and brewers.

because none of the teams added significantly to their offenses, the machine is just guessing that all of the cards will have better years than last year, and all of the reds and brewers will have worse years on average. with the ages of the players taken into account, that just doesn’t add up.

I think the part where they lose me is Cardinals, 800 RC. What–no injuries all season long? Berkman returns to have his best season ever? Schumacher and Freese tearing it up? I just don’t follow this. I do agree that it is a tossup Reds, Cardinals, Brewers, tho’