Warning: Creating default object from empty value in /home4/derok6/public_html/chop-n-change/plugins/system/cdnforjoomla/helper.php on line 27
introducing_james_piette_our_new_writer | June | 2009 Articles

2009 Archives

Introducing James Piette, Our New Writer

on .

Our writer's contest actually had two winners. Today, we're introducing James Piette, one of the two Charlie Buckets. Later this week we'll introduce our other winner, Tom, whom many of you may know from the comment threads. I'm posting this under my name because we're waiting for the tech support guys to get him his own login and stuff. Everything below the italics is all James.

Before I start rambling on and on with numbers, claims, and cultural references, I think it is important that I introduce myself. My name is James Piette and I have been an avid Braves' fan since back in the days when Fred "Fire Dog" McGriff was just acquired and the Tomahawk Chop was still intimidating.[1] My current occupation is a 3rd year PhD student in Statistics at the Wharton School of Business, which makes me one of the worst kind of people. My focus, which regrettably is not the most respected in academia, is entirely on sports, whether it be basketball, cricket, or hockey, but my true passion is in baseball.

Somehow, I impressed Alex and Kristi[2] enough to give me a shot as a writer for Chop-n-Change. I plan to post relatively frequently focusing mainly on developing stories surrounding the Braves', players who are under/over-performing, specific issues related to the team's strengths and weaknesses, and anything in between. Due to my incredibly geeky nature, these posts will inevitably be full of tirades lambasting and praising trades, players, front offices, etc., via statistical evidence, along with a little (hopefully) witty rhetoric. I plan on devoting ample time to each post's comments, as well. If I start becoming too obscure and out-of-touch, feel free to bring me back down to Earth through your hate.

As this is my first post, I feel it is appropriate to write about the question that I most often get asked when I explain to Northeners that I am not a Sox/Yankees/Mets/Phillies fan: What chance do you give the Braves to make the playoffs? Before looking at existing computer predictions, let's take a look at the standings. At 35 - 40, Atlanta is 5.0 games back of the Phillies. While plenty of teams in baseball's history have made more impressive pushes to get into the playoffs, our chances would appear to be dwindling fast.

One might suggest we have gotten unlucky, but that does not appear to be the case. Our luck score, or our current number of wins minus the Pythagorean-formulated number of wins[3], is 0 this year, compared to last year when it was -6. If luck isn't the case, is it simply the Braves' are not good enough to win the division title?

My next step when evaluating this question is to look to Baseball Prospectus' postseason odds, a page with computer predictions of postseason odds for every team. They list three sets of probabilities for each team making the playoffs, breaking each down into the expected number of wins and losses, the actual winning percentage of each team, and the chance of being crowned the division champion or wild card. I've decided ELO-adjusted is completely out of their mind[4] and the original version is not adjusted for current roster[5], so I am sticking with the PECOTA-adjusted odds.[6]

You'll note that our current chance of getting into the playoffs is about 25.6%, which breaks down into 20.9% odds at winning the division and 4.7% probability of winning the wild card. That is pretty good, considering the lead Philadelphia has. I would have pegged it around 20% myself, but I suppose this is why I am the eternal pessimist. However, this is not the most surprising number on the page. The Braves' Pct3, or the expected winning percentage, is 0.546, which, relative to the rest of the NL, is the 2nd best such number behind the Dodgers at 0.582. That is, the Braves' have the 2nd highest expected winning percentage, a higher value than any other team in the NL East. I'll let that sink in before continuing on.

Perhaps this doesn't come as a surprise to most of you. Kudos to you. For those of you who are in my shoes banging your head against the wall every time our outfield fails, this should come as a sigh of relief. I specifically remember this wasn't the case a month ago, but with all of the injuries to the Mets and the ineptitude of Philly pitching, it does not seem like too much of a stretch. Because I am a complete nerd, I decided to see what our chances of winning the division would have been if each team in the division started with their current Pct3 (i.e. for 6/29). After running 1,000,000 simulations[7], where I consider each team to have started the season with June 29th's posted, PECOTA-adjusted, expected winning percentages and play out 162 games, I got the following results for the approximate probability of each team winning the division[8]:

Braves: 0.463
Phillies: 0.397
Mets: 0.098
Marlins: 0.042
Nationals: 0.000

To have a 46.3% probability of winning the division title is a huge accomplishment for any team. While the Phillies would still have a substantial chance of taking the crown, considering how close 0.541, the Phillies' Pct3, and 0.546 are, it is a big discrepancy. Of course, this isn't even taking into account the wild card, which I did simultaneously using the same simulations.[9] The wild card was won by the Braves' 11.6% of the time; this is fairly often when conditioning on them being eligible to win it (i.e. did not win the division).[10] Thus, according to my simulations, the Braves' playoff chances if we were to flashback in time to April 5th, but using current rosters and player progressions, then we have a 57.8% probability of making it to the playoffs.

While that whole venture into exploiting PECOTA's abilities was exhausting, it puts up a few good points that I will end with. The first being that the Braves pitching depth is great, while the Phillies and the Mets are really suffering from their respective garbage they picked up this summer. Second, the Braves are still in this playoff race. With Tim Hudson coming back soon after the All-Star break and a front office that is almost always engaging in positive dealings, I think it's safe to say that our expected winning percentage will only go up. Lastly, if you work/live with any Phillies/Mets fans, a feeling I can relate to so well being in Philadelphia, then the next time they bring up Atlanta's fortune this year, you'll know that your team is the best team, right now, in the NL East. Unless either team acquires a star player. In that case, just cry loudly and they'll leave you alone.

[1] I do not mean to say that when engaged in full ferocity and enthusiasm, the Chop doesn't send shivers down your spine. I only mean to point out that I cannot recall in the past five years, which coincidentally coincides with the end of our playoff run, a Chop chant that gave me that same devilish grin that it once did when I was 13 years old.

[2] Most likely because they were sleep deprived. Nightmarish home stands like this will do that to people.

[3] A team's Pythagorean record is calculated by a simple formula that uses only the number of runs your team scored and the number of runs they allowed.

[4] They give the Marlins a 21.8% chance of winning. Yeah, OK.

[5] This is educated speculation. I can't be sure of this, but I'm pretty confident that they do not take roster changes and injuries into "full" account.

[6] PECOTA is the projection system created by Baseball Prospectus, which gives this choice even more founding.

[7] This simulation is not perfect. I assume that the total number of wins by a given team is determined by a binomial distribution, with n equal to 162 and p equal to that team's corresponding Pct3. Since there are plenty of games played these teams, the number of wins by each team is not entirely independent of one another. Also, in the case of a tie, I flipped a coin; a reasonable assumption given how random the outcome of a single baseball game between two very good teams is. However, this simplified version does a good job at approximating the odds of each NL East team's chance at winning the division. I have posted my R code here.

[8] These results are subject to minor changes due to the nature of simulations. Oh, and great job Jim Bowden. Nothing gives your team hope like... not giving your team any hope.

[9] See same code.

[10] Conditioning on losing the division gives the Braves' an estimated 21.5% chance of winning the wild card.

You Might Like...