The Caps Specialize At Being Uneven

After crushing the Islanders 6-2 the other night, the Capitals are now the NHL leader in two important categories, Power Play Percentage and Penalty Killing Percentage.  The league average for scoring when you are in the power play advantage is 18%, which Washington is leading at 28%. Looking the other way, DC leads the NHL in preventing opponents from scoring on the advantage at 8%, with the same 18% league average.  

Controlling these key moments in the game correlates into victories, as the top ten power play % teams and 8 of the top 10 penalty killing teams currently have winning records. Only once in the past 8 seasons has a team even finished in the top three in both categories at the end of the year, and that team had the best record in the league and made the Cup Finals.

Then why are the Caps only 8-7? Well, when playing at even numbers, they are below average.  Washington has outscored opponents for the season 50-42, but when at even strength they are being outscored 37-33. They are the only team in the NHL with that oddity.

DC may be relying too much on the power play since 35% of their goals are from being up a man, third highest rank in the NHL, when the average is 23%. Keep in mind that teams only average 4 power plays or 8 penalty minutes per a 60 minute game. This means that the Caps are only in each of these situations 12% of the time on the ice, while stinking up the other 76%. How about some more 5 on 5 drills at practice, coach?

2-0 NFL Teams vs. 0-2 Teams

After two weeks of the NFL season, there are eight teams that are 2-0 and eight teams that are 0-2.  I have seen the statistic several times today stating that 63% of teams that start 2-0 make the playoffs, while only 12% of 0-2 teams do. Now that 12% of the season is already done, which of the 2-0 teams should start printing out playoff tickets and which of the 0-2 teams should start working on their draft board?

First take a look at the eight teams that are off to a fast start, in multiple categories:

Since statistically only 63%, or 5 of these 8 will make the playoffs, which ones will advance? What is the best indicator of team success? The easiest trend is if the team made the playoffs last year, then they should make it again. However, half the teams on this list did not make it last year. Maybe last year was a fluke and the five year records will be a better predictor. Well Teams D and B have great percentages, while C and H have had several losing seasons recently. If “defense wins championships”, then certainly Teams C and H are in since they are among the tops in defensive yards and points allowed. The NFL is an offensive league now though, so the top offenses of Teams F and G are strong bets to make the playoffs. Game by game, winning the turnover battle is key, therefore Team C and H are starting the season off well, while Team A, B, and F need some work to do. After looking at these metrics overall, Teams C and G are playoff bound, while Teams A, B, and E will be the 3 in the 37% that do not make it.

Now for the teams on the other end of the spectrum, which started off 0-2:









Looking at the same metrics as before, Teams I and K made the playoffs last year and have above average scoring offenses, so maybe they can turn their seasons around. Team L and M are even in the turnover battle, and have above average scoring defenses. Teams J and O have great records in the past five years, so maybe they just got off to a rough start this season. Looking at the rankings overall, it looks like Team L has the best chance of being the one team to make the playoffs after a 0-2 start.

Here is the team key:



It’s not surprising that Seattle and Denver are the two best, but Chicago and New Orleans need more than history and fan base to get in. There is no guarantee Pittsburgh, NY Giants, or Washington are turning it around, while Tampa Bay’s defense and turnover margin can give their fans some hope.

Runningback’s Decline Not Just Due to Turning 30

In the world of fantasy football, there are many theories and clichés to predict a player’s rise and fall. The “third year receiver”, “handcuff” your runningback, and “avoid kickers and defenses” theories are common, and I have even written about the “30 year old RB wall” one previously. Further research goes beyond just the age of RB’s, but at carries and performance to predict their decline as well.

When looking runningback stats from the past 13 seasons, there are some interesting trends about the decline of a RB’s performance from year to year.  There are certain benchmarks signifying a good season that RB’s strive to meet year to year.

1. Since 2000, there have been 123 10 rushing touchdown seasons, but only 49 (40%) of those RB’s got at least 10 TD’s the next year.
2. Of the RB’s who were in the top #10 in scoring that year, only 47 (39%) were in the top #10 the following season.
3. 63 RB’s scored 150 fantasy points or more in a season, with only 21 (33%) reaching that mark the next year.
4. The 1,500 yard rushing mark has been eclipsed 51 times, but only repeated 13 times or 25%.

What does this mean for last year’s RB’s? Four RB’s had seasons last year that fit each of the criteria above, Adrian Peterson, Arian Foster, Marshawn Lynch, and Frank Gore.  Following the percentages above, only one or two of those RB’s will reach those benchmarks again.  Frank Gore has played seven full seasons and is already 30, and Lynch had never been over 1500 yards or 150 points before in his six year career, so they are good bets from that group to regress.

There is another factor to look at besides players peaking, which is the amount of carries. Runningbacks take a pounding carrying the ball throughout the season, and thus have shorter careers as compared to quarterbacks or wide receivers.  This wear and tear leads to decreasing point production the following year.

1. Since 2000, there have been 21 RB’s that carried the ball 370+ times in a season, and 19 (90%) of them had a decrease in points the next season for an average of 40% less points.
2. 34 RB’s with 350+ carries, with 28 (82%) decreasing an average of 36%.
3. 49 340+ carries, with 41 (84%) decreasing an average of 50%.
4. 104 300+ carries, with 73 (70%) decreasing an average of 48%.

As far as last year’s RB’s go, Foster and Peterson had 370+ carries, while Lynch had 340+ and Gore had 300+. Ray Rice had 340+ carries and met some of the earlier criteria as well. Two other RB’s got 300+ carries but were rookies or second year players, Doug Martin and Stevan Ridley.  Following the percentages, you would expect all of these RB’s, except the two young ones, to see a decrease in point totals next year.  

Therefore, when it comes to draft day, I would avoid some of these workhorses who are not likely to repeat their performances and opt for younger RB’s who are coming into their primes. 

Now It's Time to Panic for the Nats

Now that the Nationals have lost six in a row, it is officially time to panic in the nation’s capital. At this point last season, Washington was 20 games above .500, at 60-40. Now through 100 games this season, the Nat’s are four games below .500 with only a 48-52 record.  What happened to the team that was picked to make the World Series?

The Nationals' strength last year was the excellent pitching staff from top to bottom. Last season they were 1^st in the NL in ERA, opponent’s batting average, and walks/hits per inning. Now those numbers have slipped to 6^th, 7^th, and 5^th. These numbers are reflected in the bullpen as well, which went from 3^rd best to 8^th.  Even though the stats declined from last year, the numbers are still respectable, and indicative of a winning ballclub. What about the other factors in baseball?

In DC’s first place season last year, the team finished 4^th in batting average, 5^th in runs, and 6^th in on base percentage.  Now the Nationals’ bats have struggled to 13^th, 14^th, and 14^th out of the 15 NL teams in those respective categories.  They have not fared any better in the field as Washington fell from 4^th to 15^th in number of fielding errors and 2^nd to 15^th in fielding percentage.

Despite these numbers, teams find a way to win by cliche sports terms like lucky or clutch. Batters can be clutch when runners are in scoring position or when called to pinch hit. Last season they were 7^th and 1^st in batting average in those situations, but now 11^th and 12^th.  In close games, the team has faltered as well. They went from 27-21 in one run games and 13-7 in extra inning games last year to just 16-14 and 5-7 in those situations now.

                The reason the Nationals had such lofty expectations this season was that they either improved or returned each spot of their roster. The team has not seen major injuries, and is healthier overall compared to last season.  How could a team (not full of aging vets, but young rising stars) get worse without major injuries or key player movement?

                The answer? Mojo. Curses. Karma. Whatever you want to call it, but there is definitely something going on in the mental makeup of the players. It started in the last month of the regular season last year when Strasburg was cautiously benched due to his innings limit, starting the Strasburg curse. Then in one of the last home games of the year, winless Racing President Teddy Roosevelt finally won, creating the Teddy curse.  Now a mediocre president, Taft, joined the race, bringing mediocrity to the 4^th inning entertainment and possibly another curse.

On a more realistic note, one player change in the offseason might have an unforeseen mental impact as well. Goofball slugger Michael Morse brought humor to the clubhouse and broke up the tension last season with his antics, but was traded away in January. Now the pressure of meeting lofty expectations may be getting to the young team.  Perhaps the Nationals can overcome the statistic and curses to use the remaining 62 games to catch the injury-riddled Braves and win the division again.

Where’s the Love and Runs for Strasburg?

Last night, Nationals pitcher Stephen Strasburg threw seven shutout innings, lowering his earned run average to 2.24 which is the 5th best in the National League. However, Washington did not score any runs either, and ended up losing the game 4-0. Unfortunately, this has become a common occurrence this season, as Strasburg has received a major league low 2.44 runs per game from the offense, causing him to only have a 4-6 record.

With the All-Star game coming up, this lack of run support will hurt the ace’s chances to make the team.  Unfortunately win/loss record is a prominent, overrated statistic used to evaluate pitchers despite the reliance on outside factors.  If you were a manager choosing the NL pitchers, which three would you pick from the players below?

Pitcher A – Record of 10-3, team is 49-33 and is one of the most popular/successful teams
Pitcher B – Record of 9-4, 13-year veteran and former all-star
Pitcher C – Record of 9-6, team is 49-34 and is in first place
Pitcher D – Record of 4-6, team is playing .500 and under-achieving
Pitcher E – Record of 5-4, on a team with the worst record and attendance in the NL
Pitcher F – Record of 5-6, team is 11 games below .500 and in fourth place.

Clearly managers, using the antiquated notion of helping your team win is the same as a win/loss record, would choose pitchers A, B, and C. Meanwhile, there are plenty of other stats to choose from to evaluate pitchers, including earned run average (ERA), opponents batting average (BAA), and walks/hits per innings pitched (WHIP). Having read MoneyBall recently, I have even more profound appreciation for more efficient and revealing stats, like wins above replacement (WAR) and defense-independent ERA (DIPS).  Now look how the pitchers above rank using the five metrics mentioned:

Pitcher A – ERA of 3.75 (31st in the NL out of 32 qualifying pitchers), 25th in WAR, 22nd in WHIP
Pitcher B – Dead last in DIPS, WAR, and WHIP, 30th in ERA
Pitcher C – Not in the top 26 in any category.
Pitcher D – 5th in ERA, top 10 in three other categories
Pitcher E – 3rd in BAA, top 11 in three others
Pitcher F – 4th in BAA, top 13 in three others

It would now make more sense to choose pitchers D, E, and F to the All-Star game using these stats.  When looking at the averages of all these rankings, these three pitchers would rank in the top 10 overall in the NL, worthy of one of the nine starting pitcher spots on the team. Meanwhile pitchers A, B, and C are ranked 22nd, 31st, and 32nd.

Will pitchers D (Strasburg), E, and F make the team? We will find out Saturday when the rosters are revealed.

NCAA March Madness Bracket Tips (2011 Edition)

Last year I wrote a post around March Madness time about displaying some interesting trends about the percentage of seeds advancing in the tournament. I originally based the stats off of the past eight seasons. Since then I went back and included the past 25 seasons as well as the results that followed the trends last year.

In the first round:
· #12 seeds upset the #5 seeds 14 of 32 possible times, or 44% (25 season average is 34%, last year it was 25% with #12 Cornell beating #5 Temple)
· 22% of #4 seeds go down (25 year-21%, last year #4 Vanderbilt was upset)
· Meanwhile only 6% of #3 seeds lose their first game (25 year is higher at 15%, last year #3 Georgetown screwed up my bracket in the opening game)

In the second round:
· More #5 seeds make the sweet 16 than #4 seeds, (15 to 10) (25 year range is closer with #4 seeds making it 43-36 times, but last year backed it up with #5 Butler and #5 MSU advancing while only one #4 seed Purdue made it)
· Only 3 (or 9%) of #1 seeds were upset (25 year is slightly higher at 12%, which #1 Kansas’s stunner to Northern Iowa proves can be even higher)
· Meanwhile 38% of #2 seeds have been upset by the #7/#10 winner (25 year average is at 36%, and #2 Villanova’s loss to St. Mary’s was the wrong #2 I picked to fall)
· Only 5 times (or 16%) has a region gone #1, #2, #3, #4 (Which did not happen again last year)

Sweet 16 round:
· 78% of #1 seeds make the elite eight (The 25 year average is a little less at 73% and was only 50% when #1 Syracuse joined Kansas at home early as well last year)
· On the other side of the region, 78% are either #2 or #3 seeds (71% for 25 year average and 75% last year)

Elite Eight round:
· Only one time in 30 years of the 64 team bracket has all four #1 seeds made the Final Four (Which was no exception last year)
· 44% of #1 seeds make the Final Four (The 25 year average is also at 44%, while only #1 Duke made it last year)
· #2 seeds make it 25% of the time (22% over 25 years, and 25% last year with only #2 West Virginia surviving their region)
· If you add up all the seeds that make the final four, the average total is 9.25. Meaning if you pick all #1 seeds, that’s less than half the average, or if you throw in a #11 George Mason with a #3, #2, and #1, that’s double the average. (The 25 year average is 9.88, and last year’s average of 13 (#1, #2 #5, #5) was slightly above average)

While last year followed these patterns pretty closely, it does not mean I picked the correct upsets and did not win my pool. Boo. Maybe I should listen to this blog which analyzes travel time and preseason polling.

New Defensive Rankings

Who has is the best defense in the NFL? The team with the least yards allowed? The team with the least points allowed? What about the team with the most turnovers? Sacks? In any article mentioning defensive rankings, the default is yardage, followed by points. Even in 2000, when the Ravens set a record for least points allowed and regarded as one of the best defenses of all time, the Titans were the “#1 defense” that season. Allowing the least amount of points seems like the most logical choice to help your team win. So why the disparity? I couldn’t find the grand conspiracy, so I made my own rankings.

The Steelers and the Chargers are tops in points and yards allowed, but what about the other factors? I ranked the top defenses in points allowed, yards allowed, 3rd down conversion %, penalty yardage, sacks, and turnovers. I even threw in subtracted a point for each TD scored by the defense, the most direct contribution to team victories. Here is the chart showing the combined rankings (click on the picture to actually be able to read it):

San Diego tops the list due in part to being the top 11 in each of the categories. #2 Pittsburgh also is in all top 11, including top’s in turnovers as well, coming in second overall. The #3 Cowboys were #1 in both 3rd down conversion and penalty yardage, but only caused three turnovers in their three games, contributing to their 1-2 start. #3 Titans lead the sacks with two other teams and are in the top 14 in four other categories. Some of the outliers are that #16 Baltimore is #2 in yards and #5 in points, but gave up the most penalty yardage in the league and only have two turnovers. Meanwhile Buffalo has committed the least amount of penalties, but is dead last on this list since they are 27th or worse in the other five categories. If you look at the conditional formatting colors, you can see the correlation between the categories, and the standings on the left. This is only based off a quarter of the season, curious to see what it looks like at the end.