I already told you what I saw in the crystal ball for the American League in 2007. Now it’s time to talk about what may transpire on the senior circuit. Specifically, will pigs fly in the year of the boar?
Author’s Note 1: I originally published this article at Newsvine.com last Wednesday. After having watched it languish there for a general lack of interest in baseball, I decided I’d give it a shot here. I hope everyone’s cool with that.
Author’s Note 2: In an earlier article and I laid out the method I used to generate these predictions. In the interests of not forcing new readers to go back to that article – though it’s so compelling they’ll surely want to visit it again and again – I’m re-presenting a fair amount of that content here as well. However, I was kind enough to place all the “how” stuff at the end of this article. So read on. And then read further on if you’re so inclined.
Like love and marriage. Like a horse and carriage. You can’t have the American League without the National League. It doesn’t rhyme, but it’s true. And that means you can’t have some wild-eyed goon on Newsvine making mad prophesies for one without expecting to see him do the same for the other. So here we are. Once again, I went about this as scientifically as I could bring myself to, using past performance as a limited indicator of future success and all that. I also did my damnedest to keep my personal biases out of the picture. But enough about that, let’s get to it.
On With The Show!
So here they are, my predictions for the 2007 National League final standings:
NL East
New York Mets (96-66)
Philadelphia Phillies (88-74)
Florida Marlins (86-76)
Atlanta Braves (86-76)
Washington Nationals (76-86)
The Mets blow away everyone else in the East. It’s a good year in Queens, what with Willie Randolph leading the team to the best record in the major leagues and all. Heck, this division looks like a cakewalk for them, despite the fact that the Phillies, Marlins and Braves will all put up winning records. Sadly, the same can’t be said for the Nationals. Of course, it’s not like anyone in Washington is winning much these days, so this really shouldn’t be a surprise. “First in war, last in the NL East” indeed.
NL Central
Chicago Cubs (90-72)
St. Louis Cardinals (88-74)
Milwaukee Brewers (87-75)
Houston Astros (86-76)
Cincinnati Reds (81-81)
Pittsburgh Pirates (80-82)
The Cubs take the central. Paging anyone who thinks Alfonso Soriano was a bad signing for 2007. No one? Ok. Good. Now, I’ll happily admit that I was first in line to call the contract questionable for its entire length. But for 2007, at least, it sure looks like a good move. The Cardinals, Brewers and Astros look to duke it out for 2nd place and a possible shot at the NL Wild Card. Meanwhile, the Reds and Pirates, well… we can’t keep living in the 70s forever. Apologies to Brut aftershave, Chevy Chevelles and Pete Rose’s hair. Don’t worry, though. If the Cubs wind up winning the Series then the apocalypse happens and your suffering is at an end.
NL West
San Diego Padres (92-70)
Los Angeles Dodgers (87-75)
Arizona Diamondbacks (87-75)
San Francisco Giants (84-78)
Colorado Rockies (80-82)
The Padres win the west. It ain’t no tie this year. The Dodgers young guns are good, and so are the kids in Arizona. But the Padres pitching will likely be the factor that separates the wheat from the chaff out west. Despite the “best” efforts of the steroidally enhanced and eventually indicted, the Giants won’t be a factor. Meanwhile, once again the Lord will turn his back on the true believers we call the Rockies.
So, What’s Next?
Keep your eyes peeled for my next piece, in which I will prognosticate in relation to the 2007 playoffs and World Series. The astute among you have already figured out the matchups based on my predictions. But I’ve got news for you, friend – don’t bet the farm just yet.
How I Got To My Predictions
Right, so, in case you missed it before, here’s the method behind the madness. As I mentioned in the AL Predictions article, this may be more interesting to some readers that the predictions themselves….
I took the PECOTA projections for Runs Scored for each of the likely batters of each team, including bench players and the like, and did my own little proprietary mumbo-jumbo to get a reasonable team Runs Scored (RS) value. Then I worked even more math – which is likely to turn out to have been somewhat faulty, we’ll see in October – to develop a team Runs Allowed (RA) value, accounting for the likely starters, relievers and closers for each team. Then, armed with my PECOTA-based RS and RA for each team, I turned to the big man, Bill James, and his Pythagorean Record concept.
What’s that? You don’t know what the heck the Pythagorean Record thing is all about? Well, the bottom line is that you can come pretty close to projecting a team’s winning percentage with the following formula: (RS^2)/(RS^2+RA^2). One you have that percentage, all you have to do is apply it to the 162 game schedule and, voila!, you have the number of wins that team should rack up in the season. And, of course, a little simple subtraction will get you their losses, too. And that means, you’ve got their record.
Geeky Stat Geek Note
I actually used 1.83 as my exponent, because some folks have found it to project a little more accurately. Actually, some other people have found a way to determine an even more accurate exponent, but hey, despite the nature of this article I actually do have a real life that needs tending to from time to time.
Additional Information On The Process
Perhaps the most important thing to keep in mind after reading this article is that the Pythagorean Record doesn’t account for things like luck, so it’s not infallible. Along with that, one has to understand that my approach made limited concessions to the effects of injuries on the teams. PECOTA does a fair job of accounting for past injuries in determining playing time for the next season. And playing time, in the form of plate appearances, directly affects the number of runs scored by each player. So there is some consideration there, but probably not enough once hindsight kicks in later in the year.
My selection of batters and pitchers to use for RS and RA was based on the depth charts for each team as currently projected by MLB.com. I tried to include prospects who are likely to make their respective big league teams (i.e., Alex Gordon in KC). But there’s always a chance that someone comes out of nowhere during spring training.
Special NL Consideration: For the National League teams, which don’t use a Designated Hitter, I had to do something to account for the runs that would be scored by the pitcher’s spot in the lineup. The semi-arbitrary number I assigned to that slot – 25 – seemed to push the win totals too high for each team by roughly 2 wins. So, to keep things in line with the general tenor of previous seasons’ records, I adjusted the records you see above down by those 25 runs. So, in the end, these projections don’t take into account whatever relatively small number of runs might be scored by each team’s pitching staff. I’ll try to do a better job of that next year. Meanwhile, consider bumping the records by a game in favor of each team. Or don’t. That’s your call.
One last note: For each player, I used the weighted mean PECOTA values. That means that there’s little chance I accounted for a breakout season – or an absolute collapse – from anyone, regardless of whether such an occurrence is likely or not. PECOTA does include breakout and collapse chance information, but I didn’t consult that data for this project. Perhaps I will for 2008.
8 replies on “MLB 2007: National League Standings Predicted Much Sooner Than Necessary”
spelling you spelled alfonso soriano’s name wrong. you had it as Alphonso
I would have liked to see… your two “season predictions” articles combined into one…
it would have saved a little bit of re-reading.
Also, some HTML work would have made the article much easier to absorb- maybe making the headings bold or italics or something to sectionalize it a little more.
As far as voting, one way or the other… I’m still trying to absorb everything. I’ll get back to you.
fixed Fixed, thanks.
good point It’s a solid point about combining the two. They originally ran on different days, in an attempt to build up to the playoffs piece, so that’s the reason for their current bisected existence.
As far as the sectionalizing, are you not seeing any bolding in the piece? Or are you suggesting a further level of headings in addition to what’s already there? I ask only because I do have some in there that I can see, and I want to be sure it’s visible to you as well.
Thanks!
I just realized… my computer was freaking out for some reason and not showing pages on the internet correctly…
Somehow its been fixed, and I like the way you have sectionalized it.
Not to question your methodology but you have the American league with 1177 wins and 1095 losses which might make sense it the American league beat up on the National league in inter league play, but the National league is even worse with 1374 wins and 1218 loses. 🙂
You’re Right
It’s one thing I didn’t normalize in my projections – correcting for the total number of games played in the season. I really did just take the numbers I came up with for Runs Scored and Runs Allowed and run them through the Pythagorean Record formula. That is, I just list the record that each team would be expected to have based on their Runs Scored and Runs Allowed.
A lot of times teams over- or under-perform on those records. Luck, injuries, career seasons from players, etc. all influence the outcomes. And that, of course, is why they play the games. What happens on the field is far more important than any stats wonkery that a doofus with a computer can come up with.
All that said, it’s a very important consideration and one that I’ll put in my “make it better next year” list.
Thanks for reading. And always feel free to question my methodology. If there’s something I’ve overlooked or flat-out ignored I should be reminded so I can take it into account in the future.
but this isn’t just a little off — you have THREE teams in the NL with sub .500 records. That would just indicate to me that something is particularly out of whack with your math. Using stats is fine and all (I have a degree in math), but when you are consistently overpredicting wins (and your notes says that you thought you had underpredicted wins) its hard to really take the math you used seriously.