Cy Young and MVP – Predicting the Winners

Brian Walton looks at two formula-based award predictors to assess Albert Pujols' and Chris Carpenter's chances of taking home the hardware.

I have probably just about beaten the Most Valuable Player and Cy Young analysis into the ground in recent days. Yet, there is one more area that I believe bears review.

Some have correlated the individual performances of players competing for the award over time with the actual voting results with the intent to develop and refine a formula, an algorithm, that could be used to forecast the chances of competitors winning the award in any given year.

One is readily available and updated daily online, while the other is less accessible, but simple to calculate manually.

First of all, the father of sabermetrics, Bill James, and his researchers developed a way to predict Cy Young winners through a combination of innings pitched, earned runs, strikeouts, wins and losses, saves and shutouts plus a premium for playing on a division winner.

You can go to ESPN to view the details, but here is a summary of the current (as of Sunday morning) National League Cy Young candidates based on that formula.

ESPN NL Cy Young Predictor as of 8/21/05

Rank

Pitcher

Points

1

Chris Carpenter, StL

180.9

2

Dontrelle Willis, Fla

142.7

3

Chad Cordero, Was

137.6

4

Roger Clemens, Hou

134.4

5

Roy Oswalt, Hou

123.5

6

Billy Wagner, Phi

123.2

7

Pedro Martinez, NYM

120.7

8

John Smoltz, Atl

117.7

9

A.J. Burnett, Fla

116.6

10

Mark Mulder, StL

113

Cy predictor link

Without totally dissecting the formula, it is worth noting that over half of Chris Carpenter's 180.9 points come from his win/loss record. That also explains why Roger Clemens is so far behind in the scoring.

I don't know about you, but I find it difficult to fathom that Clemens will actually come in fourth in the voting, but I have to admit that I like to see Carpenter with a 38 point lead.

I have been unable to locate a similar automated predictor tool for Most Valuable Player. However, Jonathan Bernstein and Chaim Bloom at Baseball Prospectus developed a formula several years ago, then stepped back to debate its merits in detail in a three-part article series.

I am not going to argue in favor or against this algorithm, but instead present it here for your consideration. When all was said and done, Bloom concluded with this statement:

"On balance, I think the predictor is still a useful tool. It defines the normal way that NL MVPs are selected; it is the other awards that are in need of explanation. While their American League colleagues increasingly vote in seemingly random ways (Ichiro!?), National League voters appear to apply consistent standards, at least most of the time, at least when the standards yield plausible results."

BP's formula in Bloom's own words:

"Specifically, my NL MVP predictor assigns one point each for the following:

• Leading the league in runs batted in.
• Leading the league in batting average.
• Leading the league in home runs.
• Driving in 100 runs.
• Having a .300 batting average.
• Playing for a division- or league-winning team.
• Playing an up-the-middle position for a division or league winner.

Total those points for all players; those with the most points are the candidates. If there's more than one candidate, sum their batting average, RBIs, and home runs, and add a bonus for playing an up-the-middle position and/or playing for a division winner. That's it. That's going to give you your MVP."

So, let's take it out for a spin.

Baseball Prospectus' National League MVP Predictor as of 8/21/05

Candidates

RBI lead

Avg. lead

HR lead

100 RBI

.300 BA

Div. winner

Middle position

Total points

Albert Pujols

1

 

 

1

1

1

 

4

Derrek Lee

 

1

 

1

1

 

 

3

Andruw Jones

 

 

1

1

 

1

 

3

Miguel Cabrera

 

 

 

1

1

 

 

2

Morgan Ensberg

 

 

 

1

 

 

 

1

Obviously, in the case of leaders, I assumed whoever is ahead today will remain there at the end of the season. In the case of RBI, I projected the season-ending total for each player.

With six weeks of play remaining, the results of this formula can and will change. It is obvious that NL category leadership will provide the swing votes for this approach.

However, as of now, using this formula, Albert Pujols is forecast as the National League MVP winner. Being content with the findings of both tools, I think I will stop right here.

BP Report Part Three (note links at top of page to Parts One and Two)

Brian Walton can be reached via email at brwalton@earthlink.net.


The Cardinal Nation Top Stories