Skip to content

Jim Delany’s Mulligan: Legends and Leaders

I have updated this blog post, since its original publication.  So, returning readers may want to give it a second look, while first-timers may want to skim the comments for clues as to why the post was updated and, perhaps more interestingly, to find more data!

B1G Commissioner, "Uncle" Jim Delany

We all could use a mulligan, from time-to-time. Jim Delany’s should be for signing off on the dumbest division names imaginable – Legends and Leaders – after borrowing Tom DeLay’s Sharpie to draw the Big Ten’s new map. Mine will be for overlooking an odd NCAA rule, when I wrote my first blog post, which requires that, in order to have a college football conference championship game, a conference must have a minimum of 12 teams, split into 2 divisions.

I’ll let Football Outsiders expose the folly of having the 84th-best team in college football (UCLA) playing in a conference championship game, in a world where only 1 of the 6 FBS conference championship games will feature the top 2 teams in the conference, as measured by Football Outsiders’ F/+ rating.

You may recall that the Legends and Leaders were built with competitive balance as the primary goal, and traditional rivalries and geography as secondary goals.   Let’s turn that on its head and use geography and the preservation of rivalries, as primary goals, to realign the Big Ten (B1G) in a way that restores all trophy games and traditional rivalries, while satisfying that goofy NCAA rule and achieving competitive balance to the extent that that is even practical.  In doing so, I will show that competitive balance is the pot of gold at the end of the rainbow–a false promise and a fool’s errand.

Geography

Step One: Divide the 12 schools into 3 geographically sensible pods, as I proposed in Big Ten Tic-Tac-Toe.

Schools within a given pod will play the other 3 schools in their pod, annually.  This alignment preserves all of the permanent matches, which were played from 1993, when Penn State joined the B1G, until 2010, the year prior to Nebraska joining.  Here’s the map:

Step Two: Pair-up the pod members, with some consideration towards their recent, relative success (i.e. try not to pair up the two best pod members), in preparation for [reluctantly] forming 2 divisions to satisfy that NCAA rule.

These pod pairings are the only way [that I am aware of] of preserving all of the B1G’s traditional rivalries, upon completion of step 3, below.  But, before we complete step 3, I want to point out that the 3-pod alignment preserves 10 of the 14 B1G “trophy” games–all but the following:

Minnesota vs. Michigan [Little Brown Jug, 1903]
Illinois vs. Ohio State [Illibuck, 1925]
Indiana vs. Michigan State [Old Brass Spittoon, 1950]
Minnesota vs. Penn State [Governor’s Victory Bell, 1993]

Note: Ohio State vs. Michigan and Ohio State vs. Penn State are not played for trophies, but are listed among the trophy games, on Wikipedia.  I have included these two matches among the 14 trophy games–now 15, with the addition of the Heroes Trophy, for the annual Iowa-Nebraska match.

Here’s a map, which shows that more work (i.e. step 3, below) is necessary to preserve the 4 trophy games that would not be saved by a straight-pod alignment:

This map also shows that additional work (i.e. protected matches) would have to be done to preserve all trophy games, if forming two divisions based upon an east-west or north-south split, since any such split would sever the ties that are the trophy games.

Step Three: Build two, 6-team divisions out of the pod pairs.

Notice that the schools that are connected by red, dotted lines, on the previous map, are now included in the same division (i.e. are shaded the same color).

With tongue planted firmly in cheek, here are your new B1G divisions:

Loons Division:
Minnesota
Wisconsin
Northwestern
Purdue
Michigan
Penn State

Lunkers Division:
Nebraska
Iowa
Illinois
Indiana
Michigan State
Ohio State

Schools would play all of their division opponents, annually.  The result will be 5 intradivision games + 2 intrapod games [against the two pod members that a team wasn’t paired with] + 2 extra games [one each against 2 of the 4 remaining schools], once a 9th conference game is added, in 2017.  This will ensure that each school plays all of the other schools in the conference, in at least 5 years out of 10.

Rivalries

The Leaders-and-Legends format sacrificed 3 of the conference’s 14 “trophy” games and 1 of the 11 permanent matches that have been in place since Penn State joined (1993):

Note: The percentages listed in brackets are the frequencies, with which the given matches were played, during the period between the additions of Penn State and Nebraska (i.e.  1993-2010), followed by the frequencies under the Leaders/Legends format (i.e. 2011-???).

Iowa vs. Wisconsin [Heartland Trophy]: 86 games, since 1894 [100% to 40%]
Northwestern vs. Purdue [no trophy]: 78 games, since 1895 [100% to 40%]
Michigan St. vs. Penn St. [Land Grant Trophy]: 28 games, since 1914 [100% to 40%]
Minnesota vs. Penn St. [Governor’s Victory Bell]: 12 games, since 1993 [67% to 40%]

The Loons-and-Lunkers would restore all of the above to annual matches!

Before we move on, I think that it is enlightening to peak behind the curtain on the B1G machinations of power that led to the spilling of Heartland blood at the feet of the false icon of competitive balance.  Thankfully, Barry Alvarez has loose lips:

I would have loved to have protected Iowa…But there had to be some give-and-take in the process, and everybody had to give up something in order to fill the needs of people along the way.

Were these “people along the way” straw men, created by Barry to blur the appearance that he might’ve willingly traded perennial bouts with Iowa, in exchange for a one night stand with his alma mater?  Regardless, Barry keeps talking:

Never heard one scenario that would have included us with Iowa…If there was any way possible that we could have protected Iowa we would have tried to get it done, and Gary felt the same way.

I write this blog partly for fun, but mostly for self-agrandizement, of course.  But, even I will concede that I can’t be the only occasional blogger who discovered a scenario that would’ve saved the Iowa-Wisconsin rivalry.  Oh, and Gary is Barry’s counterpart, at Iowa.  Here’s Gary Barta, U. of Iowa Athletic Director, as reported by HawkCentral.com:

One of the things we knew going in was that when you add a 12th team (Nebraska), things are going to change.  And we knew it wouldn’t be perfect

Why not?  Why stop short of perfection?

Competitive Balance

Let’s return to Barry Alvarez (from Madison.com):

Ohio State separated itself from the others in the top six who were pretty close as far as won-lost records…So it had to break down along the lines of three and three — three of the top six in each division — and that was our No. 1 criteria in realignment. To make sure we achieved that priority we each gave up something.

There are 3 Loons (Penn State, Michigan, and Wisconsin) and 3 Lunkers (Ohio State, Nebraska, and Iowa), in the top-6 in overall winning percentage, during the 1993-2009 period that Alvarez was referring to.  So, Loons-and-Lunkers has that covered, too.  Here’s how it looks, graphically:

Looks pretty balanced to me.  But, lest we follow Delany’s lead, let us recall that competitive balance is a fool’s errand.

Football Outsiders provides us with its Frameau Efficiency Index (FEI), which goes back to 2007 (5 seasons * 12 B1G football teams = 60 team-seasons).  [I’d prefer their F/+, but it only goes back 3 years.]  I am going to use FEI as the best indicator of the quality of a given team.  A team’s cumulative winning percentage over the 17 seasons immediately previous to an FEI rating produced an R-squared value of 0.338.  In other words, approximately 1/3 of a B1G football team’s efficiency at doing the things necessary to win games, in a given season, can be predicted by the success rate that the team had at winning football games, in the previous 17 seasons.  That is not a strong correlation.  It’s not worthless, but it’s nothing that intelligent people should be using as the primary factor behind the alignment of a college football conference.

Here’s another graph to show how the 17-year periods that preceded a given season predicted individual teams’ winning percentages for that season, from 1910-2011 (1,211 team-seasons worth of data):

Again, the R-squared value is low; just 22% of a B1G team’s ability to win football games, in a given season, can be predicted by their ability to win football games in the previous 17 seasons.

We can also compare the cumulative, overall winning percentages of the various division alignments:

If that looks like a jumbled mess, its because it is.  A blue or red line exactly on the x-axis (middle of the graph, y = 0) would mean that the two divisions in a given alignment had identical, cumulative winning percentages, in a given season. There doesn’t appear to be any meaningful difference between the combined histories of the Leaders-and-Legends and the Loons-and-Lunkers.  In other words, it does not appear as though we could have called either alignment more or less competitively balanced than the other, if we look at the above graph and the hard, historical data:

[Almost] Since the Beginning of Time…

Cumulative Win%, 1900-2011 (Std.Dev)
Loons: .586 (.085)
Lunkers: .586 (.080)
Difference: .001

Legends: .576 (.071)
Leaders: .596 (.089)
Difference: .020

So, the schools in the Loons division and the schools in the Lunkers division have almost identical, cumulative winning percentages, since 1900, while the difference between the Legends and Leaders was twenty times less balanced!  Meanwhile, the cumulative winning percentage among the Legends’ schools was least highly variable (i.e. lowest standard deviation) among the 4 divisional alignments shown above, while the Leaders’ schools was most highly variable and the Loons’ schools and Lunkers’ schools were in between.  In other words, it is difficult to say that either alignment would have been significantly more balanced, if the B1G had been operating under these divisional alignments, in parallel universes, since 1900.  And, if forced to choose the more balanced alignment, one would have to choose the Loons-and-Lunkers.

Avg. Annual Difference in Win%, 1900-2011 (Std.Dev.)

Loons vs. Lunkers = .098 (.068)
Legends vs. Leaders = .092 (.072)

Finishing up with the data since 1900, this last set of data shows that the average annual difference between the cumulative winning percentage of one division versus the other, in a given alignment, is slightly greater for the Loons-and-Lunkers, but that the annual difference between the Legends and Leaders divisions is slightly more highly variable.  In other words, this is further evidence that neither alignment would have been significantly more balanced than the other.

The Penn State Era [1993-2010]…

Here are the same data sets, for the Penn State Era (i.e. 1993-2011).

Cumulative Win%, 1993-2011 (Std.Dev.)
Loons: .588 (.050)
Lunkers: .576 (.057)
Difference: .012

Legends: .577 (.062)
Leaders: .596 (.058)
Difference: .011

I’d be inclined to give a slight edge to Loons-and-Lunkers, if looking at the data from the Penn State Era, which is what the B1G focused on, because the standard deviation (i.e. variability) of the annual, cumulative winning percentages of the schools in the Loons and Lunkers divisions is less than the Legends and Leaders divisions.

Avg. Annual Difference in Win%, 1993-2011 (Std.Dev.)

Loons vs. Lunkers = .085 (.052)
Legends vs. Leaders = .091 (.068)

This actually may be the most compelling piece of data that suggests a slight edge to the Loons-and-Lunkers, in terms of competitive balance during the Penn State Era.  The difference between the cumulative winning percentage of the Loons’ schools and the Lunkers’ schools was both smaller, on average, and less highly variable than that between the Legends’ schools and Leaders’ schools.

Conclusions on Competitive Balance:

Individual schools’ winning percentages cannot be predicted with any degree of certainty by their past winning percentages.  And, the historic, cumulative winning percentages of my Loons-and-Lunkers suggests that they are just as balanced, if not more so, than the B1G’s Legends-and-Leaders.

The story is not much different, if we look at the average point differential or normalized point differential (i.e. points for minus points against, divided by total points), instead of Win%, or if we look at the predictive power of 1, 5, or 10 years of data on the subsequent 1, 5, 10, or 17-year periods of performance.  [Read the comments, if you want more data.]

One Caveat:

Increasing the number of protected rivalries does lend itself towards some scheduling-induced, imbalance.  For example, the most balanced schedule would include a round robin format, where every team plays every other team, each season.  That ain’t gonna happen.  Adding protected matches opens the door to some imbalance, because not all protected opponents are created equally.  The Loons-and-Lunkers includes 7 perennial matches per school, while the Legends-and-Leaders has 6.  Here are the average difficulties of schedule, per school, where 1.00 represents  a neutral schedule (i.e. 1.00 is equivalent to playing a full, round robin schedule):

Legends-and-Leaders
1.018 – Michigan
1.017 – Minnesota
1.013 – Penn State
1.003 – Nebraska
1.002 – Indiana
1.001 – Purdue
0.995 – Illinois
0.992 – Ohio State
0.987 – Iowa
0.977 – Northwestern
0.974 – Wisconsin
0.970 – Michigan State

Std.Dev. = .016

Loons-and-Lunkers (incl. net change from Legends-and-Leaders)
Note: Negative numbers indicate that a school’s schedule would become more difficult, if switching from Legends-and-Leaders to Loons-and-Lunkers.
1.040 – Minnesota (-.024)
1.028 – Wisconsin (-.055)
1.024 – Michigan State (-.054)
1.023 – Michigan (-.005)
1.023 – Penn State (-.010)
1.010 – Ohio State (-.018)
0.978 – Iowa (+.009)
0.971 – Indiana (+.031)
0.969 – Illinois (+.026)
0.968 – Nebraska (+.035)
0.942 – Northwestern (+.035)
0.939 – Purdue (+.061)

Std.Dev. = .034

So, Minnesota’s schedule, in the Loons-and-Lunkers alignment is 4% more difficult than it would be, if the B1G were playing a full, round robin, and 2.4% more difficult than their average Legends-and-Leaders schedule.  Now, there may be some who think that this is not insignificant–apparently, the B1G powers-that-be would argue so.  Legends-and-Leaders produces a more balanced schedule, per school, if given an 8-game schedule.  However, the move to a 9-game B1G schedule will balance these numbers out, especially if the Leaders-and-Legends format uses the 9th game to create additional protected matches.  For example, a 9th-game in the Loons-and-Lunkers alignment would allow Minnesota and Wisconsin to double their chance of playing Illinois and Indiana, in a given season, while doubling the chance that Northwestern and Purdue get games with Ohio State and Nebraska.

C’mon, Jim!

I can accept that separating the Big Ten’s four, traditional, football powers – Ohio State, Michigan, Penn State, and Nebraska – might not be a bad idea, if only to maximize the chance that 1 or 2 of the B1G’s most valuable assets are filling up primetime pixels, on the night of the B1G’s most valuable showcase event.  As a Badgers’ fan, it has already been fun to watch the middling folks from Madison and East Lansing foil The Man’s master plan.  But, it’s unfortunate that the contortions of the powerful sacrificed a handful of rivalries that some fans actually care about…much more so than what the name of their school’s football division is or, quite probably, which division their school is actually in.  Uncle Jim just needed to rearrange his priorities.

Braun + Fielder vs. Yount + Molitor vs. Aaron + Mathews

I spent the summer wondering – sometimes aloud – if the heart of the Milwaukee Brewers’ lineup featured the best one-two combination in Milwaukee baseball history.  Actually, I was pretty confident that it was, at least in terms of offensive production.  Now, as Brewers’ fans celebrate Ryan Braun’s MVP award and prepare to say goodbye to Prince Fielder – and with the healing effects of the 5+ weeks that separate fans from their team’s playoff ouster – it is time to compare the reigns of Ryan Braun and Prince Fielder, to those of Robin Yount and Paul Molitor, and Hank Aaron and Eddie Mathews.

Ground Rules

First, the older couples (Yount/Molitor and Aaron/Mathews) played together, as teammates, for much longer than the 5 years of Braun and Fielder that Brewers’ fans enjoyed.  So, I will limit my comparison to the 5-year periods of the older couples’ co-existence, where they were of similar ages to Braun and Fielder.  Second, Braun and Fielder are the same age, in baseball terms–they both just completed their age-27 seasons.  And, since baseball players’ careers tend to peak at approximately age-27, I have limited the 5-year periods for each pair to that which ends when one player from the pair completes his age-27 season.  Here are the periods of comparison:

Study Period: Player 1 (age range) + Player 2 (age range)
2007-2011: Braun (23-27) + Fielder (23-27)
*1979-1983: Yount (23-27) + Molitor (22-26)
1955-1959: Aaron (21-25) + Mathews (23-27)

* 1981 was a strike-shortened, 109-game season.  I have extrapolated Yount’s and Molitor’s statistics over a full, 162-game season.

By this methodology, the advantage should rest with those pairs who had more older years in the study period, since they are closer to the theoretical, age-27 peak.  In other words, Braun and Fielder should have a slight edge over Yount and Molitor, since Molitor didn’t make it to his age-27 season in this study period, and Yount and Molitor should have a slight edge over Aaron and Mathews, since Aaron fell 2 years short of the age-27 theoretical peak.

Going to WAR

I used an average of three Wins Above Replacement (WAR) metrics to calculate total value: Baseball Prospectus’s Wins Above Replacement Player (WARP), Baseball Reference’s WAR (rWAR), and Fangraph’s WAR (fWAR).  Fangraphs provides a good overview of WAR.  The important thing to note is that WAR considers offensive run production and defensive run prevention.  [Note: WARP and rWAR consider baserunning; fWAR does not.]  The consideration of defense places added value on positions that are more difficult to play (i.e. SS>2B>CF>3B>RF>LF>1B>DH).  Here’s how the pairs compare, over the course of the 5-year periods of study:

Aaron + Mathews = 74.5 wins (7.5 wins per player-season)
Yount + Molitor = 56.6 wins (5.7 wins per player-season)
Braun + Fielder = 44.1 wins (4.4 wins per player-season)

Wow!  Aaron and Mathews were pretty good.  They were also consistent, durable, and very good at young ages.  Hank Aaron was a 5-win player (i.e. All-Star-caliber) by the age of 21, didn’t put up a sub-5-win season until age-38, and put up 4 consecutive, 9+-win seasons (i.e. MVP-caliber), immediately after our study period ended!  Eddie Mathews was also nearly a 9-win player, at age-21, finishing 2nd in the MVP voting (1953), and didn’t drop below 5-wins until age-32 (1964)–a dozen, consecutive years of All-Star caliber performance!  For comparison, Braun and Fielder each produced two, 5+-win seasons (2009 and 2011), including Braun’s recently completed, 7.3-win, MVP season.  Molitor also accumulated two, 5+-win seasons during our study period, while Yount recorded four.  Here are the 30 individual seasons from our study period:

WAR – Player-Age (Year)
10.7 – Yount-26 (1982) [MVP]
9.2 – Aaron-25 (1959)
8.5 – Mathews-27 (1957)
8.2 – Aaron-23 (1957) [MVP]
7.9 – Aaron-24 (1958)
7.8 – Mathews-23 (1955)
7.4 – Mathews-25 (1957)
7.3 – Braun-27 (2011) [MVP]
7.2 – Aaron-22 (1956)
7.1 – Yount-27 (1983)
7.0 – Yount-25 (1981)
7.0 – Yount-24 (1980)
6.7 – Molitor-25 (1982)
6.7 – Mathews-26 (1958)
6.1 – Aaron-21 (1955)
5.7 – Fielder-25 (2009)
5.7 – Molitor-22 (1979)
5.6 – Mathews-24 (1956)
5.4 – Braun-25 (2009)
5.3 – Fielder-27 (2011)
4.7 – Braun-26 (2010)
4.4 – Fielder-23 (2007)
4.3 – Molitor-26 (1983)
3.6 – Molitor-23 (1980)
3.5 – Braun-24 (2008)
3.1 – Fielder-26 (2010)
2.8 – Molitor-24 (1981)
2.7 – Yount-23 (1979)
2.7 – Braun-23 (2007)
2.0 – Fielder-24 (2008)

And, here are the best tandem seasons from our study period:

WAR – Teammates (Year)
17.8 – Aaron-Mathews (1959)
17.4 – Yount-Molitor (1982)
15.6 – Aaron-Mathews (1957)
14.6 – Aaron-Mathews (1958)
13.8 – Aaron-Mathews (1955)
12.7 – Aaron-Mathews (1956)
12.6 – Braun-Fielder (2011)
11.4 – Yount-Molitor (1983)
11.2 – Braun-Fielder (2009)
10.6 – Yount-Molitor (1980)
8.5 – Yount-Molitor (1979)
7.7 – Braun-Fielder (2010)
7.0 – Braun-Fielder (2007)
5.9 – Yount-Molitor (1981)
5.5 – Braun-Fielder (2008)

Glove Hurts

So, let’s ignore defense.  Baseball Prospectus offers True Average (TAv):

TAV considers batting as well as baserunning, but not the value of a position player’s defense. The TAv adjusted for all-time also has a correction for league difficulty. The scale is deliberately set to approximate that of batting average. League average TAv is always equal to .260.

Aaron + Mathews = .336
Braun + Fielder = .314
Yount + Molitor = .293

So, the old Braves are the runaway winners; Yount’s value, as a shortstop, puts him in the discussion for MVP among these six players; and the defensive shortcomings of Braun and Fielder muffle their offensive value.

Favre vs. Rodgers, Addendum

I exchanged emails with a Favre adorer, today, who informed me that I can’t make comparisons of quarterbacks from different eras, as I did in “Favre vs. Rodgers, Objectively.”  Here’s what I wrote in that email:

The best thing about the metric that I used (DVOA) is that it is comparing a quarterback to his contemporaries. It is saying that the degree to which Rodgers is better than the other NFL QB’s, today, is so dramatically greater than the degree to which Favre was better than the other QB’s of his day, as to be too significant to ignore and dismiss with claims that it is “due to the era or the rules.”

Favre was 29% better than average, in 1995, which was his best rating, as a Packer (4th overall). Rodgers was 34% better than average, last year (4th), and he is 60% better, this year (1st).

Favre never finished a season ranked #1 in DVOA. So, Favre was never the best QB in the league, per pass attempt! And, despite his durability, the best he ever did in total value (DYAR) was tied for first, with 33-year-old, Vinny Testaverde, in 1996–35-year-old, Dan Marino, finished 3rd. So, Favre, in his prime, was barely better than those immobile, old men!?!

And, even as a Favre fan, there was no doubt that Barry Sanders (2,053 yds; 6.1 yds/attempt!; 14 total TD) was a more deserving MVP, in 1997, than Favre, who led the league in TD passes, but didn’t even complete 60% of his passes, had a 35:16 TD:INT ratio, and a pedestrian 92.6 QB rating. Favre was 9th in the league in value per pass attempt (14.5% DVOA), in 1997, and 4th in total value, among QB’s (936 DYAR).

Together, with the benefit of hindsight and advanced statistics – and Aaron Rodgers’ right arm – let us reconsider The Favre Myth.

When 162 Isn’t Enough

This is just a simple, I concur post…

Image

It looks like Major League Baseball is going to add one wild card team and have the two wild card teams per league play a one-game playoff to advance to the Divisional Series round.  Why not 3-game series, with all 3 games played at the home fields of the wild card teams with the best record?  Both my 3-division and 5-division integrations of the American and National Leagues (i.e. equal numbers of AL and NL teams, per division) incorporated the 3-game wild card series.  Joel Sherman, from the New York Post, agrees.

Favre vs. Rodgers, Objectively

Tom Silverstein’s Brett Favre vs. Aaron Rodgers blog post on the Milwaukee Journal-Sentinel’s website received it’s first comment at 10:31 AM, on November 17, 2011.  Within 20 hours, it had received 510 comments!  Packers’ fans get excited about a lot of things.  Rodgers-or-Favre might still be the most volatile.

Subjectively Objective Manner

Silverstein sides with Favre:

…Rodgers is playing better than any quarterback in the NFL right now, but I hesitate to say that he’s playing at a level higher than Favre ever did.

Here’s why: he’s not.

He then fires a preemptive shot across the bow at anyone who might think otherwise, effectively insulting his readers before providing any factual evidence to support his claim:

My guess is that many of those who want to say Rodgers is playing better than Favre never saw Favre play in his prime. Either that or they’re just incapable of viewing this argument in an objective manner.

Objective?  Great idea!

Having a gut feeling that Favre’s numbers would stand up to Rodgers’ at some point during his career, I went back and looked at the ’95 and ’96 seasons, when he was roughly the same age as Rodgers is now, and was in the midst of winning three straight MVP awards.

“Gut feeling?”  Sounds objective.

Silverstein searched the vaults and found a 16-game stretch in Favre’s past that is comparable to Rodgers’ current 16-game stretch.  While we’re at it, let’s compare The Rolling Stones’ greatest hits to side 2 of Yellow Submarine and the first half of Abbey Road.  [To be fair, cherry picking Favre’s best 16-game stretch, as a Packer, probably isn’t wholly inappropriate, as a best defense against the is-Rodgers-better-than-Favre-ever-was question.  It just seems a contorted way of settling the issue.]  Here’s what Silverstein’s subjectively objective manner produced:

Favre: 68.3% completions, 49 TD to 7 INT, 32 sacks, and a 119.1 passer rating, while his team went 14-2 overall, and 2-1 in the playoffs

Rodgers: 70.8% completions, 42 TD to 7 INT, 37 sacks, and a 121.5 passer rating, while his team went 15-0, and 4-0 in the playoffs

Setting aside the fact that Rodgers wins the rate stats (completion percentage and passer rating), thus suggesting that, if any conclusion should be drawn, it is that Rodgers would be the choice, if you had but one pass to be made; and the fact that Rodgers didn’t lose during his current 16-game stretch; and the fact that we don’t yet know if this is the high water mark of Rodgers’ career – that we’ll never see him play better over any future 16-game stretch – there’s a better, more objective way to compare the quarterbacks…

My Stats are Better Than Your Stats

Football Outsiders gives us DVOA, or Defense-adjusted Value Over Average. “This number represents value, per play, over an average quarterback in the same game situations. The more positive the DVOA rating, the better the player’s performance.” In simpler terms, “DVOA means a quarterback with more value per play.”  Let’s look at the entirety of Aaron Rodgers’ career and compare it to Favre’s gun-slinging, at the same age:

DVOA (League Rank) – QB (Age) – Awards:

15.7% (9) – Favre (25)
14.8% (14) – Rodgers (25)

29.2% (4) – Favre (26) – MVP
22.7% (9) – Rodgers (26)

33.8% (4) – Rodgers (27) – SB MVP
14.7% (4) – Favre (27) – MVP

59.6% (1) – Rodgers (28) – MVP?
14.5% (9) – Favre (28) – MVP

Favre wins the comparison of age-25 and age-26 seasons.  However, Favre had the benefit of 29 NFL starts under his belt, as he began his age-25 season; Rodgers had zero.  Regardless, Favre never had a season in a Packers’ uniform like Rodgers’ age-27 season.  And, Favre never had a season that was even remotely close to Rodgers’ partial age-28 season.  To put Rodgers’ 2011 season in perspective, he would have to throw 679 passes (i.e. about 19 games worth of passes) at the success rate of his worst season (14.8%, in his inaugural starting campaign) to bring his current DVOA (59.6% after 322 passes) down to Favre’s best season as a Packer (29.2%).  To put it another way, Rodgers’ would have to be 10% worse than an average quarterback (-9.9% DVOA, to be exact), over his final 7 games, for his current 59.6% DVOA to drop all the way down to Favre’s level (29.2%).

Best Season Ever (since 1992)

Here are all of the 40%+ DVOA seasons on record (i.e. since 1992):

[Min. 300 passes]
60.6% – P.Manning (2004)
59.6% – A.Rodgers (2011)
56.9% – T.Brady (2007)
53.3% – T.Brady (2010)
51.0% – P.Manning (2006)
45.9% – P.Rivers (2009)
44.2% – T.Brady (2009)
42.9% – R.Cunningham (1998)
41.0% – D.Brees (2009)
40.6% – P.Manning (2007)
40.5% – P.Manning (2005)
40.5% – C.Pennington (2002)
40.4% – S.Young (1994)

So, forget Rodgers vs. Favre; let’s talk about Rodgers vs. history!  Rodgers has a shot at Peyton Manning’s all-time DVOA record, while Favre doesn’t even make the list.

Best QB Ever (since 1992)

Finally, here are a select few quarterbacks, ranked by DVOA, during their age-25 through age-28 seasons, with the familiar [flawed] passer rating added for reference:

DVOA – Quarterback…Passer Rating (Rank among this group)

29.4% – Peyton Manning…97.5 (2)
29.2% – Aaron Rodgers…104.4 (1)
26.9% – Philip Rivers…96.3 (3)
21.5% – Drew Brees…94.1 (6)
19.3% – Ben Roethlisberger…95.2 (4)
18.6% – Brett Favre…94.7 (5)
18.0% – Tom Brady…88.9 (7)

A Higher Level

There, that’s objective.  And, unlike Silverstein, I do not hesitate to say that Rodgers is playing at a higher level than Favre ever did.

Here’s why: he is.

The MLB Three-way

I suspect that many a blogger suffers more from a lack of time than from a lack of ideas. I definitely fall into the former category and the anticipation of becoming a father for a second time, this winter, includes the understanding that I will not be gaining free time in the coming year. So, while I sit on a spreadsheet that has the potential to change how the world looks at last year’s Big Ten realignment, last weekend’s leak of potential MLB realignment got me thinking…

If I Were King

First, I prefer David Pinto’s 5-division, 6-team integration of the National and American Leagues, which I improved upon in Bud Selig’s Curtain Call, for reasons of symmetry, geography, and intrigue:

Cliff’s Notes Version of Bud Selig’s Curtain Call:

Each of the 5 divisions would have 3 NL and 3 AL teams, which would require that Arizona (or Houston) and Washington shift to the AL, and Minnesota shifts to the NL.  Teams would play 18 games versus their 5 intradivision rivals and 3 games versus all other teams.  Games at AL stadiums would use the DH, such that AL teams would play 117 out of 162 games with the DH, while NL teams would play the same number of games without the DH.  The 5 division winners would receive first-round playoff byes, while 6 wild cards would play 3-game series, entirely at the home stadium of the team with the better record.  On the plus side, this would greatly reduce travel, it is geographically sensible, and it produces perfectly symmetrical schedules, with the season beginning and ending with home-and-home series versus all 5 division opponents. The biggest drawbacks are the need for 3 teams to agree to switch leagues, the marginalization of the DH, which, presumably, the Players Association would balk at, and the reduction of the Cardinals-Cubs rivalry to a thrice per annum trip down memory lane.

Thrice as Nice

My latest plan combines Pinto’s AL-NL integration concept, with a 3-division, 10-team alignment that satisfies geographical sensibilities and requires only one team – the Houston Astros – to switch leagues. The Astros have been rumored to be one of the franchises under consideration for moving from the NL to the AL, to balance the leagues at 15 apiece. So, this is perhaps a more likely realignment option:

The Schedule

The schedule would include one, 3-game series versus each of the teams from the other divisions, and 11 or 12 games versus the 9 teams from within a team’s division. In other words, teams will visit their intradivision rivals twice per season, and will visit all other teams once every other year.

60 Interdivision Games (20 teams * 3 games/team)

102 Intradivision Games (6 teams * 11 games/team + 3 teams * 12 games/team)

162 Total Games

The primary scheduling benefit of this plan, as compared to the rumored realignment to two, 15-team leagues, is that there would not have to be interdivision play every night and the schedule could be constructed such that teams would be playing their intradivision rivals at the end of the season.  Additionally, travel times and costs would be reduced, since teams would play a mere 30 games at opponents who are not within their geographically aligned divisions.

The Playoffs

If forced to accept this plan – as opposed to the five, 6-team divisions that I prefer – and if forced to select a playoff format, I would go with one wild card and 7-game series through the semifinals and World Series.  However, that would not accomplish MLB’s objective to increase the playoff field from the current total of 8 teams.  Five wild cards would leave the total number of playoff teams at 8.  Ten wild cards would allow for a 3-game, wild card round, with the 5 winners advancing to join the 3 division winners, in what would amount to a continuation of the current playoff format–5-game quarterfinals series, 7-game semifinals, and 7-game World Series.  I would allow the 5 best wild card teams to host all 3 games of the wild card series, as a reward for the better record and as a means of making it difficult to finish with the 13th-best record and make a playoff run to a World Series title.  This format would reward the 3 division winners with a first round bye, which would make the pennant races meaningful, unlike the current format, where a wild card offers no significant playoff disadvantage vis-a-vis a division crown.

Brewers’ Defense: Roadblock To A World Series? | FanGraphs Baseball

http://www.fangraphs.com/blogs/index.php/brewers-defense/

Pitching Pennies: Zack Greinke

As my brother Tim would say, “Wowzerz!”

The Summary

The reported Brewers’ blockbuster of Alcides Escobar, Lorenzo Cain, Jeremy Jeffress, and Jake Odorizzi, for Zack Greinke and Yuniesky Betancourt, makes Milwaukee legitimate NL Central contenders, but at a significant cost to their future.  And, while this Sports Illustrated cover may, in fact, be correct in its pronouncement; and, while I have no knowledge of the availability of other alternatives, such as Ricky Nolasco, Chad Billingsley, and James Shields; I am certain that the Brewers could’ve made themselves contenders at a much lower cost.  Surely, the Marlins would’ve listened to an offer for Nolasco, if Cain and one of Jeffress or Odorizzi were dangled.  And, I can’t imagine that the Dodgers and Rays would not have paused, if Mat Gamel and one of the two pitchers were offered for Billingsley or Shields.

The Means

This is a haul for the Royals:

Alcides Escobar is the prize.  He was the Brewers #1 prospect, heading into the 2010 season, and #12 in all of baseball, according to Baseball America.  Escobar’s best skill is his defense, and Baseball America gave him a 70 score, on the 20-80 scouting scale, for his glove.  Escobar finished 2010 with a +4.7 UZR/150, which was good enough for 8th among all Major League Baseball shortstops.  He accomplished this as a 23-year-old rookie, mind you.  For comparison, Yuniesky Betancourt (28 years old) was a -9.2 (18th overall), in 2010, and has a -8.3 UZR/150, over the course of his career.  With the bat, Escobar finished 2010 with a .232 TAv (.260 is league-average), while Betancourt hit .249 TAv (.244 career).  So, they both make more outs and produce fewer runs than the average major leaguer.  But, Escobar has the benefit of the growth curve in front of him, while Betancourt is what he is: a below-average offensive shortstop, who hurts you in the field, too.  Bill James projects Escobar to hit .272/.318/.364 (AVG/OBP/SLG), in 2011, while Betancourt is projected to hit .264/.293/.397.  Has Craig Counsell (career +7.0 UZR/150; projected .237/.332/.308) been signed, yet?  If not Counsell, is there anything left on the farm that would net us this guy, and his career +3.0 UZR/150 and projected .286/.339/.434?

Jose Reyes

Lorenzo Cain was the Brewers’ #8 prospect, heading into 2010, and very much looked the part of a major league centerfielder.  Cain hit for a .288 TAv (.306/.348/.415), in 158 plate appearances, as a 24-year-old.  He also played the outfield with abandon.  Only a dozen major league outfielders outplayed his +9.3 UZR/150, in 2010.  Bill James projects Cain for a .279/.345/.379 line.  Cain’s exodus leaves fellow-25-year-old, Carlos Gomez, and his better glove / lesser stick (career +14.5 UZR/150 and .240 TAv; projected .258/.311/.363), as the incumbent.

Jeremy “Cheech” Jeffress – nickname courtesy of @ProfessorParks, relating to the pitcher’s marijuana record – was the Brewers #4 prospect (#100 overall), heading into 2009.  Jeffress rebounded from a 100-game suspension for drugs; made it to the bigs, in 2010; flashed his potential to miss bats (7.2 SO/9) and the plate (5.4 BB/9), in 10 MLB innings; and touched triple-digits, on national TV, in the recently completed Arizona Fall League.

Jake Odorizzi was the Brewers top prospect, at the time of the trade, according to Tom Haudricourt, who ranks Brewers prospects for Baseball America (Twitter feed):

I do Top 30 prospects for Baseball America. Had Lawrie first, then moved Odorizzi to No. 1 when Lawrie was dealt. Now, both are gone.

And, now, Odorizzi, who blossomed in 2010, with a 10.1 SO/9 and 3.0 BB/9 performance, in 120 2/3 innings, in Low-A ball, will be lucky to be a top-10 prospect, in everybody’s #1 farm system–that of the Kansas City Royals.

For that kind of loot, I have to wonder how close the Brewers might’ve been to acquiring  “The Mexicutioner,” too.

The Ends

Alas, this is what Brewers’ fans will have to look forward to–and, it is the “ends” that Doug Melvin is hoping will justify the “means”:

Bill James’s projected ERA – Pitcher (age) – Years of Team Control:

3.57 – Zack Greinke (27) – 2
3.60 – Yovani Gallardo (25) – 5
3.77 – Shaun Marcum (29) – 2
4.05 – Randy Wolf (34) – 3
4.45 – Chris Narveson (29) – 5

And, here’s the updated Brewers’ rotation, sorted by 2008-10 SIERA:

3.36 – Greinke
3.53 – Gallardo
3.78 – Marcum
4.03 – Narveson
4.31 – Wolf

Finally, here are the data for the other pitchers that I’ve mentioned, juxtaposed with Greinke:

Bill James’s projected ERA – 2008-10 SIERA – Pitcher (age) – Years of Team Control:

3.57 – 3.36 – Greinke (27) – 2
3.63 – 3.74 – Billingsley (26) – 2
3.92 – 3.26 – Nolasco (28) – 2
4.06 – 3.75 – Shields (29) – 2

The Lament

The Brewers got the best guy.  But, Greinke and Melvin must deliver a playoff spot or two, to justify the cost of a half-decade of certainty, at baseball’s most important, non-battery position.

Alcides Escobar, Kansas City Royals Shortstop, 2011-2015

Note: All UZR data and Bill James’s projections were taken from FanGraphs.com.

Pitching Pennies: Ricky Nolasco

This Tom Haudricourt blog post hardly qualifies as a trade rumor, but its worth taking a closer look at the subject:

Ricky Nolasco

The Florida Marlins would have little interest in my favorite Brewers’ trade chip, Mat Gamel, since they have Gaby Sanchez at 1B, Logan Morrison in LF, Mike Stanton in RF, and Matt Dominguez ready to play 3B.  The Marlins could use a true CF, despite their insistence that Chris Coghlan can handle it.  Perhaps they would be willing to trade Ricky Nolasco, and his 2 years of team control, for Carlos Gomez, his 3 years of team control, plus a prospect.  I presume that the Brewers would balk at including Lorenzo Cain, but Nolasco just might be the pitcher that you give up Cain for:

2008-10 SIERA – Pitcher (Age) – Years of Team Control:
3.26 – Ricky Nolasco (28) – 2
3.53 – Yovani Gallardo (25) – 5
3.78 – Shawn Marcum (29) – 2
4.03 – Chris Narveson (29) – 5
4.31 – Randy Wolf (34) – 3

And, there you have it, contenders!

P.S. For comparison:

3.36 – Zack Greinke (27) – 2
4.27 – Joe Blanton (30) – 2
4.58 – Fausto Carmona (27) – 4

Pitching Pennies: A Prelude

There’s an outside chance that I might find time to do some more work on this topic, before it becomes old news.  In the meantime, some raw numbers for my fellow Brewers’ fans to chew on:

Our Guys…

2008-10 SIERA – Pitcher (Age) – Years of Team Control:

3.53 – Yovani Gallardo (25) – 5
3.78 – Shawn Marcum (29) – 2
4.03 – Chris Narveson (29) – 5
4.31 – Randy Wolf (34) – 3

You will find a mere 2 pitchers on this entire post with better numbers than “Yo.”

Their Guys…

2008-10 SIERA – Pitcher (Age) – Years of Team Control – Cost:
3.74 – Chad Billingsley (26) – 2 – One year of Prince Fielder for Billingsley + 2 years of James Loney?
3.75 – James Shields (29) – 2 – Six years of Mat Gamel for Shields?
4.16 – Matt Garza (27) – 3 – Six years of Gamel for Garza?
4.16 – Derek Holland (24) – 6? – Fielder + a prospect for Holland + 4 years of Chris Davis?

Holland was my idea, based solely on this.  That one would be intriguing, but I doubt that the Rangers would bite.  The others are all rumors that have surfaced at one time or another, to which I’ve applied some spin.  I’d prefer one of the Rays – Shields or Garza – over anything involving Fielder.

Free Agents…

2008-10 SIERA – Pitcher (Age) – 2008-10 IP:
3.13 – Brandon Webb (32) – 231
3.44 – Cliff Lee (32) – 667
3.57 – Rich Harden (29) – 381
3.97 – Chris Capuano (32) – 66
4.13 – Carl Pavano (35) – 455
4.37 – Jeff Francis (30) – 248
4.40 – Chien-Ming Wang (31) – 137
4.51 – Kevin Millwood (36) – 558
4.54 – John Maine (30) – 261
4.55 – Brad Penny (33) – 324
4.57 – Freddy Garcia (35) – 228
4.59 – Doug Davis (35) – 388
4.61 – Dave Bush (31) – 474
4.76 – Bruce Chen (34) – 203
4.82 – Jarrod Washburn (36) – 330
4.83 – Nate Robertson (333) – 320
4.85 – Chris Young (32) – 198
5.09 – Ryan Rowland-Smith (28) – 324
5.14 – Jeff Suppan (36) – 441

Could the Brewers get two of Webb, Harden, and Capuano, for the price of one Pavano?  Heck, could they get all 3?

Career: 138-143, 4.69 ERA, $58,125,000.00