Sunday, March 21, 2010
MIT Sloan Sports Analytics Conference Feature Panel
What Geeks Don't Get: The Limits of Moneyball
The panel features Mark Cuban, Jonathan Kraft, Daryl Morey, Bill Polian, and Bill Simmons with Michael Lewis as the Moderator. Enjoy, folks and thanks, ESPN.
What Geeks Don't Get: The Limits of Moneyball
The panel features Mark Cuban, Jonathan Kraft, Daryl Morey, Bill Polian, and Bill Simmons with Michael Lewis as the Moderator. Enjoy, folks and thanks, ESPN.
Thursday, March 18, 2010
NOTES FROM THE MIT SLOAN SPORTS ANALYTICS CONFERENCE
It has been nearly two weeks since the Sloan Sports Analytics Conference so this blog post is late. I did tweet from the conference and you can read a particularly good recap over at baseball musings.
The Baseball Analytics panel consisted of John Abadoned of the Cardinals, John Dewan of plus/minus defensive metric fame, Dan Duquette, Shiraz Reham from Arizona, and Tom Tippett of the Red Sox. Abandoned seemed to be the most forth coming among the trio representing teams. He broke news that MLB will have a league wide injury/health electronic medical record system.
In terms of player evaluation, Abandoned still thinks that the margin for error when projecting player performance is still large and can be refined. But it seemed like he was also more concerned than Tippett or Reham about the future enhanced pitch fx system that can identify a fielder’s positioning. Abandoned thought it would make it too easy for other teams to measure defense; hence the Cards would have less of an advantage.
Abandoned wants to measure or do a better job at evaluating “makeup” (work ethic, desire, probability of giving into temptation, etc.) The question was not asked but I would think potential acquisitions, certainly amateurs, would be given personality tests or psychology evaluations before being drafted or signed. The Red Sox may have a leg up on the Cards in this area as they have a Sports Psychologist in Bob Tewksbury.
Tippett pretty much stuck with the Yawkey Way talking points on this winter. The Red Sox did not have a change in philosophy that was now emphasizing defense. Based on their player evaluations, including defensive metrics, the market was undervaluing players whose best skills were with the glove. Plus, the Olde Towne Team was able to sign the position player free agents to relatively short contracts (a.k.a. low financial and production risk) making the transactions easy decisions. John Dewan commented that the Red Sox 2010 run differential will be 60-80 runs better than last season.
Other Baseball Analytics Notes:
* Dewan states defensive metrics can tell you about 60% of a players glove ability.
* Teams wrestle with changing a pitcher's mechanics to give them a better shot at staying healthy when it could make them less effective. It was not mentioned but all I could think was Max Scherzer getting dealt to the Tigers from the D-Backs.
* Dan Duquette’s bio says “In addition to being recognized as the primary architect of the 2004 World Championship team…”
* Duquette wants to know how to consistently develop “20 game winners.”
* Tippett defended the Red Sox so called “babying of pitchers” stating that they project playing in October where starters could be asked to make four to six intense playoff starts, which would give them a mid-30’s GS total.
Other Observations and Notes:
* Bill Polian says the “Patriots are the team of the decade.”
* Polian knew the Pats were going on 4-and-2. It was “100% the right call.”
* Mark Cuban pays for clutch and an example is Jason Kidd; Daryl Morey of the Rockets does not.
* Cuban states that NBA teams try to hide and protect their morons on the floor. Without a doubt, it was the funniest line of the conference.
* The NBA has a long way to go to catch up with MLB metrics as data collection is more difficult.
It has been nearly two weeks since the Sloan Sports Analytics Conference so this blog post is late. I did tweet from the conference and you can read a particularly good recap over at baseball musings.
The Baseball Analytics panel consisted of John Abadoned of the Cardinals, John Dewan of plus/minus defensive metric fame, Dan Duquette, Shiraz Reham from Arizona, and Tom Tippett of the Red Sox. Abandoned seemed to be the most forth coming among the trio representing teams. He broke news that MLB will have a league wide injury/health electronic medical record system.
In terms of player evaluation, Abandoned still thinks that the margin for error when projecting player performance is still large and can be refined. But it seemed like he was also more concerned than Tippett or Reham about the future enhanced pitch fx system that can identify a fielder’s positioning. Abandoned thought it would make it too easy for other teams to measure defense; hence the Cards would have less of an advantage.
Abandoned wants to measure or do a better job at evaluating “makeup” (work ethic, desire, probability of giving into temptation, etc.) The question was not asked but I would think potential acquisitions, certainly amateurs, would be given personality tests or psychology evaluations before being drafted or signed. The Red Sox may have a leg up on the Cards in this area as they have a Sports Psychologist in Bob Tewksbury.
Tippett pretty much stuck with the Yawkey Way talking points on this winter. The Red Sox did not have a change in philosophy that was now emphasizing defense. Based on their player evaluations, including defensive metrics, the market was undervaluing players whose best skills were with the glove. Plus, the Olde Towne Team was able to sign the position player free agents to relatively short contracts (a.k.a. low financial and production risk) making the transactions easy decisions. John Dewan commented that the Red Sox 2010 run differential will be 60-80 runs better than last season.
Other Baseball Analytics Notes:
* Dewan states defensive metrics can tell you about 60% of a players glove ability.
* Teams wrestle with changing a pitcher's mechanics to give them a better shot at staying healthy when it could make them less effective. It was not mentioned but all I could think was Max Scherzer getting dealt to the Tigers from the D-Backs.
* Dan Duquette’s bio says “In addition to being recognized as the primary architect of the 2004 World Championship team…”
* Duquette wants to know how to consistently develop “20 game winners.”
* Tippett defended the Red Sox so called “babying of pitchers” stating that they project playing in October where starters could be asked to make four to six intense playoff starts, which would give them a mid-30’s GS total.
Other Observations and Notes:
* Bill Polian says the “Patriots are the team of the decade.”
* Polian knew the Pats were going on 4-and-2. It was “100% the right call.”
* Mark Cuban pays for clutch and an example is Jason Kidd; Daryl Morey of the Rockets does not.
* Cuban states that NBA teams try to hide and protect their morons on the floor. Without a doubt, it was the funniest line of the conference.
* The NBA has a long way to go to catch up with MLB metrics as data collection is more difficult.
Tuesday, March 02, 2010
2010 A.L. EAST FORECAST
Baseball Prospectus is forecasting the MFY at 90 wins, the Sox at 95 and TB with 93 wins. BP forecasts runs scored and against based primarily on their PECOTA player projections, then plugs that into Bill James Pythagorean formula to calculate winning percentage. James’ Pythagorean equation has a 9% margin for error mostly due blow out, one-run and other close games. Besides the aforementioned run distribution, player forecasts, head-to-head record and the potential for additions to the organization can impact the projections. Run distribution and head-to-head record is too difficult, if even possible, to plan for and significantly impact. The baseball gods have more say on these two factors than any player, manager or general manager, which leaves three factors to analyze.
Player Projections
PECOTA does a good job at projecting player production, but injuries, odd career and young players tend to give the system issues. With that in mind, the Rays forecast has the most risk. It is projecting that Ben Zobrist is for real after a 2009 breakout season with only a significant dip in BA. More importantly, PECOTA has the young trio of T-Bay Rays pitchers (David Price, Jeff Niemann, Wade Davis) each throwing around 150 innings with an ERA slightly above 4.00 and 2-to-1 K:BB ratios. One or even two out of three is reasonable but hitting on 100% is highly unlikely.
The MFY’s PECOTA player projections appear to be dead on with one exception – Javy Vazquez. It is forecasting that Vazquez’s K:BB to improve over his three years with the White Sox. It is likely considering his great season in Atlanta while also putting less impact on it since he is in his mid-30s. But PECOTA may not adjust a change to the A.L. East from the N.L.; only League-to-League. Also, forecasting the impact of MFY Stadium is volatile given only one year of data.
Beyond the big two, Josh Beckett and Jon Lester, the Sox rotation has question marks to varying degrees. PECOTA appears to be optimist on three. John Lackey has made 27 and 24 starts in his last two seasons. So in his thirties, it is unlikely that he takes the ball 30 times and pitches 190 innings. Daisuke Matsuzaka is a “national treasure” and “like box of chocolates…you never know what your gonna get.” Similar to the T-Bay Rays trio of young pitchers, Clay Buchholz’s forecast is on the optimist side – ERA under 4.00 and slightly better than a 2-to-1 K:BB ratio. As with any system, PECOTA is consistent.
Trade Targets
As side from the T-Bay Rays, who are unlikely to deal with the Sox or MFY and be out of October baseball, Adam Dunn and Adrian Gonzalez are the probable impact bats on the market come July. On paper, neither slugger is a fit for the Olde Towne Team or the MFY. It is doubtful that the Rays will part with the assets necessary to acquire either big bat.
If healthy, Ben Sheets or Gil Meche would upgrade any rotation. Meche would be owed roughly $18M through 2011 at the time of a deal, so the Rays are out. He also has a limited no-trade clause. Both the Sox and MFY might find Meche more attractive as he would fill a spot in 2011 that could be vacated by Josh Beckett with Boston and Andy Pettitte or/and Javy Vazquez in NYC. Sheets at $10+M in 2010 is probably to rich for the Rays blood, but not for the Sox, MFY or other contenders. Although the prospect price will probably be too high for the A.L. bEasts.
A.L. EAST Prediction (Last year, I did pretty well. I swung and missed only twice on B.J. Upton and John Smoltz. Everything else was solid, even the trade speculation.)
SOX - 93 wins – Starters 3-6 are not as productive as PECOTA projection
MFY - 90 wins (Wild Card)
TBR - 84 wins – Crawford, Pena, Soriano, etc. are traded at the deadline
Baseball Prospectus is forecasting the MFY at 90 wins, the Sox at 95 and TB with 93 wins. BP forecasts runs scored and against based primarily on their PECOTA player projections, then plugs that into Bill James Pythagorean formula to calculate winning percentage. James’ Pythagorean equation has a 9% margin for error mostly due blow out, one-run and other close games. Besides the aforementioned run distribution, player forecasts, head-to-head record and the potential for additions to the organization can impact the projections. Run distribution and head-to-head record is too difficult, if even possible, to plan for and significantly impact. The baseball gods have more say on these two factors than any player, manager or general manager, which leaves three factors to analyze.
Player Projections
PECOTA does a good job at projecting player production, but injuries, odd career and young players tend to give the system issues. With that in mind, the Rays forecast has the most risk. It is projecting that Ben Zobrist is for real after a 2009 breakout season with only a significant dip in BA. More importantly, PECOTA has the young trio of T-Bay Rays pitchers (David Price, Jeff Niemann, Wade Davis) each throwing around 150 innings with an ERA slightly above 4.00 and 2-to-1 K:BB ratios. One or even two out of three is reasonable but hitting on 100% is highly unlikely.
The MFY’s PECOTA player projections appear to be dead on with one exception – Javy Vazquez. It is forecasting that Vazquez’s K:BB to improve over his three years with the White Sox. It is likely considering his great season in Atlanta while also putting less impact on it since he is in his mid-30s. But PECOTA may not adjust a change to the A.L. East from the N.L.; only League-to-League. Also, forecasting the impact of MFY Stadium is volatile given only one year of data.
Beyond the big two, Josh Beckett and Jon Lester, the Sox rotation has question marks to varying degrees. PECOTA appears to be optimist on three. John Lackey has made 27 and 24 starts in his last two seasons. So in his thirties, it is unlikely that he takes the ball 30 times and pitches 190 innings. Daisuke Matsuzaka is a “national treasure” and “like box of chocolates…you never know what your gonna get.” Similar to the T-Bay Rays trio of young pitchers, Clay Buchholz’s forecast is on the optimist side – ERA under 4.00 and slightly better than a 2-to-1 K:BB ratio. As with any system, PECOTA is consistent.
Trade Targets
As side from the T-Bay Rays, who are unlikely to deal with the Sox or MFY and be out of October baseball, Adam Dunn and Adrian Gonzalez are the probable impact bats on the market come July. On paper, neither slugger is a fit for the Olde Towne Team or the MFY. It is doubtful that the Rays will part with the assets necessary to acquire either big bat.
If healthy, Ben Sheets or Gil Meche would upgrade any rotation. Meche would be owed roughly $18M through 2011 at the time of a deal, so the Rays are out. He also has a limited no-trade clause. Both the Sox and MFY might find Meche more attractive as he would fill a spot in 2011 that could be vacated by Josh Beckett with Boston and Andy Pettitte or/and Javy Vazquez in NYC. Sheets at $10+M in 2010 is probably to rich for the Rays blood, but not for the Sox, MFY or other contenders. Although the prospect price will probably be too high for the A.L. bEasts.
A.L. EAST Prediction (Last year, I did pretty well. I swung and missed only twice on B.J. Upton and John Smoltz. Everything else was solid, even the trade speculation.)
SOX - 93 wins – Starters 3-6 are not as productive as PECOTA projection
MFY - 90 wins (Wild Card)
TBR - 84 wins – Crawford, Pena, Soriano, etc. are traded at the deadline
Thank you - Red Sox!
This site is not associated with the Boston Red Sox, Major League Baseball nor Major League Players Assoication
© 2003-2010 Thomas J. Fratamico, III
All Rights Reserved
