best counter

Your Ad Here
Just The Sports: 2007-11-11

Just The Sports

Friday, November 16, 2007

Imitated and Duplicated

A poorly kept secret in college football is there are very few innovators among the coaching ranks. The rest of the coaches are glorified copycats, taking note of what works for other teams and then implementing them in their own programs in the following seasons. One need not even squint to see the widespread use of the spread offense, allowing teams to put up more points and also turn their quarterbacks into even riskier professional prospects. Coaches are even known to copy ineffectual strategies, as evidenced by the puzzling popularity of unorthodox kickoffs and rugby-style punts instead of simply trusting their coverage teams. This season, the most noticeable act of plagiarizing has come from head coaches imitating what Urban Meyer and the Florida Gators did with Tim Tebow last year and utilizing their own back-up running quarterbacks in the same way.

Due to the fact the LSU Tigers are in the SEC conference with Florida, I will focus on Ryan Perrilloux and compare what he has done so far this season to what Tebow did last year on the championship Gators team. Surprisingly, since I consider Tim Tebow to be a singular talent (most likely because I have a natural affinity for left-handed athletes being left-handed myself) I predicted before I started compiling data from the play-by-play of both teams' games available on that Perrilloux would have no chance of equaling Tebow's success. However, my thinking was not exactly accurate.

As far as running the ball, Tebow has a slight edge over Perrilloux, only because he has a higher success rate (58.4% to 51.3%). Other than the fact Tebow did a better job of gaining a successful amount of yardage, Perrilloux has been just as explosive a runner as Tebow. Tebow gained 4.8 extra yards per successful run while Perrilloux has amassed 4.6 extra yards per successful run so far this season. It is also worth noting that Tebow failed by an average of 3.7 yards and Perrilloux has done so by an average of 3.5 yards.

Ryan Perrilloux knocks Tebow of the 2006 season off his pedestal a bit by having had better passing statistics in his back-up role. A lot of the credit for his superiority is due to 20/25, 298 yards, three touchdown game against Middle Tennessee when he started in lieu of Matt Flynn, LSU's regular starter. Because of that performance, Perrillou has a passing success rate of 72.5% with 11.0 extra yards per successful pass and 5.7 yards needed per failed pass. Tebow's 2006 passing stats of a 64.9% passing success rate with 12.3 yards per successful pass and 5.2 yards per failed pass are not shabby, but are no equal for what Perrilloux has done for the Bayou Bengals.

They are also equal in scoring touchdowns. Tebow scored thirteen touchdowns in fourteen games last year and Perrilloux is a little ahead of that pace with nine touchdowns in nine games played.

On the field, Perrilloux has every shot to be as good with LSU as Tebow currently is at Florida, but talent will mean nothing if he is unable to change his attitude and find a way to stay out of further trouble.

Note: Successful runs gain 40% of yardage on first down, 60% of yardage on second down, and 100% of yardage on third/fourth downs. Successful passes gain 45% of yardage on first down, 60% of yardage on second down, and 100% of yardage on third/fourth downs.


Thursday, November 15, 2007

Ohio State and USC

Since the start of this, the 21st century, the Ohio State University Buckeyes and the University of Southern California Trojans have been among the elite programs in college football, setting the standard by which other college football teams are measured, and they have done so thanks to head coaching changes they made before the 2001 season when they handed over the reins to Jim Tressel and Pete Carroll, respectively. Each program was experiencing similar difficulties when they made the hires they did. Although the Buckeyes and the Trojans had storied programs, they had each fallen on hard times, USC more than Ohio State, struggles that lasted for the first year of Tressel's and Carroll's tenures. Carroll's USC team only went 6-6 and Tressel's Buckeyes only bettered that record by one win by going 7-5.

After that first season, the teams coached by these two coaches have flourished in amazing fashion and though the Buckeyes and Trojans have yet to face each other head-to-head while Tressel and Carroll have been there to coach on the sidelines of such a contest, I wanted to see if these statistics would indicate a clearly superior team or at least one that would have a decided advantage should the aforementioned match-up actually happen.

To do so was impossible. Not only have these two teams handily dismantled their competition, but they have done so in virtually identical fashion, save for slight variations due to differing ideologies. Offensively, for the eighty-seven USC games and the eighty-six Ohio State games I have statistical data for*, USC has thrown statistically significantly more passes per game (34.7 to 25.0) and also has a statistically significant advantage in completion percentage (62.5% to 59.7%). Despite the lower completion percentage, Ohio State has a very slight, but not significant, advantage in yards per pass attempt (7.9 to 7.8) and a significant advantage in yards per catch (13.2 to 12.5), meaning that USC quarterbacks throw safer, shorter passes which bumps up their completion percentages while Ohio State quarterbacks take more chances down the field so there is no clear winner in the passing time. In the rushing game, USC has averaged 4.4 yards per rush to Ohio State's 4.2.

Defensively, USC is again significantly superior to Ohio State when it comes to completion percentage, this time because they hold their opponents to a lower completion percentage (54.4% to 57.0%), and once again the Ohio State Buckeyes make up the difference by holding their opponents to comparable yards per pass attempt (5.9 to 6.0) and yards per catch (10.3 to 11.0). Both teams are amazingly tough against the run, with Ohio State giving up 2.9 yards per rush and USC allowing 3.0 yards per rush.

Simply looking at how the teams have done on offense against only non-conference opponents eliminate the superiority Ohio State held over the Trojans in terms of yards per catch. Ohio State averages 13.1 yards per catch in non-conference opponents against USC's 12.7. Also working in USC's favor is an increase of their lead in rushing averages, averaging 4.8 yards per rush while Ohio State has been averaging 4.1 yards per rush.

Then again, when it comes to defending non-conference opponents' passing attack, the significant advantage USC held in completion percentage disappears (55.1% to 56.8%). Against the run, Ohio State holds opponents to fewer yards per rush (2.8 to 3.3).

As good as Ohio State has been against non-conference opponents, they raise their level of play against their conference opponents, offensively speaking, and they re-open the significant advantage in yards per catch (13.3 to 12.4) and USC was unable to re-open their own advantage in completion percentage (62.3% to 59.7%). Really, the teams are a wash and equally adept at moving the ball and scoring points.

Defensively against non-conference opponents, USC maintained their significant advantage in completion percentage allowed (54.0% to 57.1%), although that was the only category in which they were appreciably better than the Buckeyes.

What makes these teams so great are the fact they are equally proficient in dominating their non-conference and conference foes. No matter who they are facing, the two programs perform the same way, which for them means they are performing as two of the top programs in the country.

*Note: Statistical data does not include Ohio State's 2001 match against Purdue in which the Buckeyes defeated Purdue, 35-9.


Tuesday, November 13, 2007

Where Pitches per Plate Appearances Matters Most (Pt. II)

Last year during my exhaustive research into finding out where pitches per plate appearance matters most, I made the prediction that the correlation data I uncovered during the 2006 season probably followed the same patterns for every baseball season no matter the year or the era. With the 2007 regular season having been completed for more than a month now, I figured it was finally time to put my spreadsheet where my mouth was and see if a prediction I made turned out to be correct.

Once I inputted the last player's data and looked at all the correlation coefficients, I discovered that I had been right in making a prediction, although I could not see I was exactly surprised by the results since they only made sense. Once again, the data set with the highest positive correlation to pitches per plate appearance was a statistic I came up with last year, which was to take the difference between a player's on-base percentage and his batting average. The correlation coefficient was not as high this year as last year, but at .745, it is still nothing to sneeze out and still a remarkably high correlation once you think about the different variables that result in a hitter getting on base.

Not only was the same data set the highest both years, but the correlation coefficients for the other six data sets were in the exact same order for each year. Coming in with the second highest correlation was weighted on-base average (.274) and the other correlation coefficients ranked in this order: gross product average (.265); isolated power (.209); slugging percentage (.099); batting average (-.187). One should remember when looking at numbers that the closer the correlation coefficient is to zero then the less of a relationship the variables have with each other, the closer the correlation coefficient is to 1 the more directly proportional the variables are, and the closer the correlation coefficient is to -1, the more inversely proportional the variables in question are.

Impatient hitters, or hitters with a low number of pitches seen per plate appearance, are making themselves less productive by keeping their on-base percentages lower than it should be in three ways. The first way is these players are putting too much faith in their ability to hit safely as a way to get on base. No player whose batting average makes up most of his on-base percentage is going to be a consistent player. In addition, by swinging at balls early in the count take away the option of drawing a walk, almost as if a walk is not a manly or exciting enough way to get on base. Lastly, there is no guarantee a player is swinging at the best possible pitch for him to drive when he puts the ball into play too early, which is why isolated power (slugging average minus batting average) has a higher correlation coefficient than both slugging percentage and batting average. The more patient hitters will have the most extra base hits.

The majority of professional ballplayers in any sport are less than intellectually inclined and probably do not understand much statistical analysis so it is up to baseball teams, who are interested in making their players as best as they can be, to let their employees know how to increase productivity. Taking a lot of pitches will not improve a hitter's batting average, but it will allow him to improve their on-base percentage in relation to what his batting average has always allowed him to achieve.


Monday, November 12, 2007

Coco Crisp and Johnny Damon

When the Boston Red Sox replaced Johnny Damon with Coco Crisp before the 2006 regular season after Damon defected to the New York Yankees, the trading for and signing of Crisp made perfect sense. The Red Sox smartly refused to match either the number of years or the number of dollars Damon received from the Yankees (four-year, $52 million), fearing due to Damon's style of play that he would break down before the end of four years and be unable to earn his monstrous contract. Therefore, the Red Sox went out and found a cheaper, younger version of Johnny Damon, signing Crisp to a three-year, $15.5 million contract after he was already promised $1.75 million for the 2006 season.

Previous to the 2006 season, Johnny Damon had amassed a batting line of .290 BA/.350 OBP/.431 SLG/.265 GPA, statistics Coco Crisp had matched with his own batting line of .287 BA/.326 OBP/.424 SLG/.253 GPA and Crisp was and still is six years younger than Damon, making it more likely he would be able to sustain his level of production longer than Damon. Or so the Red Sox thought.

Unfortunately, Coco Crisp has failed to live up to expectations. Not only did Crisp only play in 104 regular-season games, but when he only hit an anemic .264 BA/.317 OBP/.385 SLG/.239 GPA, his worst hitting season since he played thirty-two games in 2002 when he was twenty-two years old. At the same time, Damon hit .285 BA/.359 OBP/.482/.282 GPA, making one wonder if the Red Sox had made a mistake in letting him go.

Then the second season of their respective tenures with their new teams began, the 2007 for those keeping count at home, something interesting occurred, not with Coco Crisp but with Johnny Damon. Crisp had another sub-par season when compared to his pre-2006 numbers, but now looks like the best he has to offer, when he only hit .268 BA/.330 OBP/.382 SLG/.244 GPA. Damon, though, experienced a drop-off in production, only hitting .270 BA/.351 OBP/.396 SLG/.257 GPA.

Factoring in each player's two seasons, there is no statistically significant difference in their batting averages, on-base percentages, slugging percentages, or gross product averages so neither team has gotten good performances from either of their outfielders or at least the production either team predicted they would receive.

If a winner has to be compared now between the Red Sox and the Yankees in this lackluster contest, it would have to go to the Red Sox for the simple fact they are devoting less payroll for Crisp's worthlessness at the plate while the Yankees are grossly overpaying for Damon. Then again, if Crisp is pushed out by Jacoby Ellsbury like the playoffs gave us a glimpse of happening, by the end of the 2009 season, maybe the Yankees will be the winners.