Saturday, November 22, 2003

The Division and Correlation of Wins Between Offense and Pitching in 2003'.


It has been a constant debate as whether a team benefits more from having a stronger offense, stronger pitching, or a balanced approach.

With numbers from 2003', I will attempt to determine the Wins Contributed (WC) by each team based on their Runs Scored for offense per game (RS) and Runs Allowed per game (RA) for pitching, given an average defense. Instead of basing it on actual wins totals for 03', I will use each team's pythagorean record (PR), which bases a team's W-L record based on RS and RA, which is ideal for this study.


Hitting: Using a method that is similar to calculate the team's PR and Offensive Winning % (OWP), I have squared RS and divided it by the sum of that team's RS and the RS for the League (LRS) (AL and NL are seperated), this is simply (RS²)/(RS² + LRS²) which will provide the win percentage of that team if that team had average pitching and defense based from their offensive production, I will call this (OWR). For example, if Boston had avg. pitching and defense, their overall winning % would have been 59.8% (.598). Next, we take that .598 and multiply it by 162 to see how many wins Boston would have had with avg. pitching and defense (OWR*162), which will give us 96.09 wins (TOW).

Pitching: Following a similar method as mentioned above, except we will use RA and the League Runs Allowed (LRA) the only changes will be that the numerator with will be the LRA instead of the team's RA and the denominator will remain the same (LRA²)/(RA² + LRA²), I will call this (PWP) . This will give us the win percentage of a team with an avg. offense and defense based from their pitching production. To use the Boston example again, if Boston had an avg. offense and defense, their overall winning % would have 48.6% (.486). Like above, we take that .486 and multiply it by 162 to see how many wins Boston would have had with an avg. offense and defense (PWP*162), which would give us .78.86 wins (POW).

Combining the two: Now that it has been determined each team's offensive and pitching win totals, we must apply it to the team's PR and then calculate what percentage of those wins were created by the offense and the pitching. This is done by adding up the wins created by the offense and the pitching (TOW+POW) & using the Boston example, it would be 96.09+78.86=174.95. Now that we have the combined win totals, we can now plug one of them to determine the winning % caused by each of them by taking the offensive win total divided by the offensive and pitching totals (TOW/ (TOW+POW)). With Boston, that would determine that the offense was responsible for 55.1% (.551) of the wins and the pitching was responsible for 44.9% (.449) of the wins in 2003'. Now using the pythagorean win total of Boston in 03' of 94 wins, we can take that .551 and multiply it by 94 to show that the Boston offense contributed to 55 wins and Boston's pitching contributed to 39 wins.

The final step will be to determine which teams were offensively dominated, pitching dominated, or balanced which will be done by assigning teams with a +.525 Offensive Wins Created % as offensive based, teams with a +.525 Pitching Wins Created % as pitching dominated, and teams within the .475-.525 range as balanced. Then, we will be able to compare which of those groups had the greater success for the 2003' season.



Offense Created Wins
St. Louis-49
San Francisco-44
San Diego-33
New York-31
Los Angeles-30

Pitching Created Wins
Los Angeles-53
San Francisco-49
St. Louis-39
New York-37
San Diego-33

Offensive Created Win %
St. Louis-.556
San Diego-.503
San Francisco-.469
New York-.465
Los Angeles-.365

Pitching Created Win % (this is just the reversal of the Offense Created Win %)
Los Angeles-.635
New York-.535
San Francisco-.531
San Diego-.497
St. Louis-.444


Offensive Wins Created
New York- 48
Kansas City-42
Tampa Bay-34

Pitching Wins Created
New York-48
Kansas City-36
Tampa Bay-34

Offensive Wins Created %
Kansas City-.540
New York-.500
Tampa Bay-.495

Pitching Wins Contributed %
Tampa Bay-.505
New York-.500
Kansas City-.460

Offense dominated NL teams:
Colorado-78 wins
St. Louis-88 wins
Atlanta-96 wins
Milwaukee-66 wins

Total Wins:391
Avg. Win Total per Team:78.2

Balanced NL teams:
San Diego-66 wins
Philadelphia-88 wins
Houston-94 wins
Florida-88 wins
Pittsburgh-76 wins
Montreal-80 wins

Total Wins:492
Avg. Win Total per Team:82

Pitching dominated NL Teams:
Chicago-85 wins
San Francisco-93 wins
Arizona-84 wins
New York-69 wins
Los Angeles-83 wins

Total Wins:494
Avg. win Total per Team:82.8

Offense dominated AL Teams:
Boston-94 wins
Toronto-87 wins
Texas-69 wins
Kansas City-78

Total Wins:328
Avg. win Total per Team:82

Balanced AL Teams:
New York-96 wins
Minnesota-85 wins
Baltimore-74 wins
Tampa Bay-68 wins
Chicago-88 wins

Total Wins:411
Avg. win Total per Team:82.2

Pitching dom. AL Teams:
Seattle-97 wins
Oakland-96 wins
Anaheim-80 wins
Cleveland-73 wins
Detroit-49 wins

Total Wins:483
Avg. win Total per Team:79

NL and AL combined:

Offense dom. Teams:
Colorado-78 wins
Boston-94 wins
St. Louis-88 wins
Toronto-87 wins
Atlanta-96 wins
Texas-69 wins
Milwaukee-66 wins
Kansas City-78 wins
Cincinnati-63 wins

Total Wins:719
Avg. Wins per Team:79.88

Balanced Teams:
San Diego-66 wins
New York (AL)-96 wins
Philadelphia-90 wins
Minnesota-85 wins
Houston-94 wins
Baltimore-74 wins
Florida-87 wins
Tampa Bay-68 wins
Pittsburgh-76 wins
Chicago (AL)-88 wins
Montreal-80 wins

Total wins:904
Avg. Total per Team:82.18

Pitching dom. Teams:
Seattle-97 wins
Chicago (NL)-85 wins
Oakland-94 wins
San Francisco-93 wins
Arizona-84 wins
Anaheim-80 wins
Los Angeles-83 wins
Cleveland-73 wins
New York (NL)-69 wins
Detroit-49 wins

Total Wins:807
Avg. Wins per Team:80.7


There is a small win total variance between teams, but pitching dominated teams had a slightly greater chance of reaching .500. The offensive dominated teams had 4 of the 9 teams above .500, the balanced teams had 6 of the 11 teams above .500, and the pitching dominated teams had 6 of the 10 teams above .500. The balanced teams had a higher Win Total per team, followed by pitching dominated teams, and then by the offense dominated teams. Of the 8 playoff teams, 2 were offensive teams, 3 were balanced, and 3 were pitching dominated. Overall, the balanced teams had the better record per team and have shown the most success in the post-season the past 5 years. Of the last 10 teams to have played in the World Series 9 of them have been balanced, including the last 8. The last team that has played in the World Series that was not balanced was Atlanta in 99'.

Sunday, November 16, 2003


I will begin with the stats that apply to both starters and relievers, then use the stats that have to be seperated by each sector. The minimum amount of innings used will be over 25IP and I have seperated the stats of Juan Cruz as a starter and as a reliever, but I did combine the stats of Juan's when it could be applied (APR, dipsERA, S%, and WS).

Adjusted Pitching Runs (APR)
The amounts of runs a pitcher prevents compared to an avg. pitcher in the same amount of innings with no park factor influence.

Prior 43.4
C. Zambrano 27.8
Wood 25.3
Borowski 12.5
Guthrie 8.8
Farnsworth 8.3
Remlinger 4.83
Clement 3.8
Veres -1.4
Wellemeyer -6.8
Alfonseca -11.4
Juan Cruz -12.0
Estes -24.5

Fielding Independent Pitching Runs/dipsERA (FIP/dipsERA)
TangoTiger's version of DIPS, similar accuracy as Voros McCracken's famous DIPS without the countless # of steps. The only difference is McCracken's did include Park Factors (which is easy to calculate), but Tango did an outstanding job and the accuracy is there with or w/out the Park Factors. DIPS or dipsERA is an estimation of how effective or ineffective the pitcher was based on the stats that are independent of defense.

Prior 2.61
Farnsworth 2.91
Borowski 3.09
C. Zambrano 3.21
Wood 3.22
Remlinger 4.08
Clement 4.29
Juan Cruz 4.58
Veres 4.76
Guthrie 4.82
Alfonseca 5.34
Estes 5.47
Wellemeyer 5.56

Strand Rate (S%)
The percent of how many allowed runners are stranded. This is not a completely accurate stat for an individual pitcher because it is dependent on other members of the staff (bullpen).

Guthrie 88%
Prior 81%
Remlinger 79%
Wood 78%
Borowski 78%
C. Zambrano 76%
Farnsworth 73%
Clement 69%
Estes 69%
Veres 65%
Wellemeyer 62%
Cruz 61%
Alfonseca 54%

Win Shares (WS) (pitching only)
Bill James' stat that credits an individual player's contrubtions to the team in terms of wins.

Prior 22.57
Wood 17.56
C. Zambrano 17.15
Borowski 13.97
Clement 10.52
Farnsworth 7.17
Remlinger 5.86
Guthrie 4.04
Veres 1.94
Alfonseca 1.11
Juan Cruz .36
Estes .28
Wellemeyer .00

Support Neutral Wins Above Replacement (SNWAR) (starters only)
Michael Wolverton's statistic on how many wins a starter would have over an avg. starting pitcher (replacement) in a neutral setting. Prior, Wood, and zambrano were among top 30 in baseball while Estes was in the bottom 10.

Prior 5.4
Wood 4.7
C. Zambrano 4
Clement 2.2
Juan Cruz 0
Estes -1.8

Game Score (G Sc)
Bill James measurement for starting pitchers.

Prior 62.8
Wood 61.2
C. Zambrano 55.8
Clement 53.8
Juan Cruz 46.2
Estes 42.6

Adjusted Runs Prevented (ARP)
Michael Wolverton stat that measures the numbers of runs that reliever prevented over an average pitcher in the same amount of innings, it is dependent on the situation when the reliever enters and leaves the game.

Borowski 14.5
Farnsworth 10.3
Veres 3.5
Guthrie 3
Remlinger 2.3
Wellemeyer -10.5
Juan Cruz -12.5
Alfonseca -13.3


1. The big 3 of Prior, Wood, and Zambrano were the key to the 03' season. They were able to compensate the poor performance of Estes and turn the rotation into the strength of the team.

2. Estes was awful, while it can be debated who insisted on signing Estes as far as who thought it was a need to have a LH starter regardless of recent past production and/or who thought that Estes could turnaround the trend of his poor performance; there is no question of how awful he was last year and how much it actually req'd of the remaining starters to balance out his inability to produce.

3. Clement had a substantial drop-off compared to 02' and while he still well above avg. for a #4 starter, it can be questioned whether the Cubs did equal value in return for his salary. He had a drop in strikeouts and a rise in HRs, which is very important for Clement who will always have a questionable BB/9 ratio.

4. Next year will be a great test to see how much an extended workload can impact a pitcher the following year. Personally, I'm like the Swiss when it comes to Pitch Counts, it does increase the chances for injury, but it does not guarantee it. Baker did have several opportunites to rest starters throughout the regular and postseason and failed to capitalize on that luxury. I am clueless as to why a manager who sees the need to rest his positional player fails to exercise the same belief with the starting rotation.

5. For a micromanager to remain successful at managin a bullpen it is essential for that team to have 4 above avg. relievers. Borowski and Farnsworth fit that bill perfectly, they were head and shoulders the best on the Cubs and should not considered for any trade possibilites b/c of the usage pattern on the bullpen.

6. Who was the best pitcher against LH? Dave Veres, who while he did not get many chances vs. LHP should not ave been ignored because his best was the split-finger which due to movement (down and away to LH) was much more effective than the LOOGY situations employed with Remlinger and Guthrie. There is a reason why Remlinger and Guthrie have been more effective against RH than LH and that is b/c of their go-to-pitch, which is the change-up. To ignore previous results b/c of a R/L match-up is a reason why they will have to go with LH who have been productive against LH and not vice versa to ensure the best percentages for success.

7. While on the subject of Veres, despite a mid-80's FB and a splitter in the low 80's he was able to prevent BBs and keep the ball in the yard while having a fair K/9 ratio. His ERA was not indicative of his overall production to the team.

8. Guthrie is the exact opposite of Veres, his ability to get by with smoke and mirrors and being able to keep a low ERA might be beneficial to his upcoming FA, but not to his potential employer. His H/IP & HR/IP were avg., his BB/K ratio were poor and he was able to survive with having Farnsworth and/or Borowski as one likely to prevent the baserunner he allowed.

9. It will be interesting to look 2 years into the future and see if the Cubs did get equal value for the quick signing of Remlinger. To sign a 37yo reliever to a long-term deal is playing with fire, but Remlinger was able to have a decent BB/K ratio and had some key outs in certain situations and despite walking too many batters and allowing way too many HRs, if used properly he plays a vital role next year.

Friday, November 14, 2003

I will begin with a somewhat Sabermetric review of 2003' w/players who had over 100ABs and finished the season with the Cubs. I will begin the review with the offensive stats 1st and then work my way to the pitching stats.

Batting Runs (BR)
Batting Runs calculates the offensive value of a player. 0 is avg, +5 would mean that player was 5 runs above avg and -5 would equate to that player being 5 runs below avg.

Sosa 32.5
Alou 19.7
Lofton 18.33 (Pitt+Chc)
Patterson 12.47
Grudzielanek 11.19
A. Ramirez 9.63 (Pitt+Chc)
Karros 7.54
Choi 3.96
Goodwin .45
Simon -.33 (Pitt+Chc)
R. Martinez -1.34
O'Leary -5.57
D. Miller -5.98
Bako -6.06
Alex S. Gonzalez -7.78

Bases per Plate Appearence (BPA)
Simply put, how many bases they avg. per plate appearence.

Sosa .574
Patterson .548
Lofton .516 (Pitt+Chc)
Choi .510
Alou .497
A. Ramirez .467 (Pitt+Chc)
Goodwin .459
Karros .449
Grudzielanek .445
Simon .443 (Pitt+Chc)
Alex S. Gonzalez .430
D. Miller .397
R. Martinez .388
Bako .387
O'Leary .373

Linear Weighted Power (LWpwr)
Pete Palmer formula that measures a hitter's power based on his ability to hit 2Bs, 3Bs, and HRs.

Sosa 51.96
Choi 44.81
Patterson 41.49
Alex S. Gonzalez 39.22
A. Ramirez 39.09 (Pitt+Chc)
Alou 38.50
Karros 33.02
Lofton 32.56 (Pitt+Chc)
Simon 32.05 (Pitt+Chc)
O'Leary 29.75
Miller 29.66
Grudzielanek 26.86
Bako 24.85
R. Martinez 22.17
Goodwin 20.06

Batting Base Performance Value (BPV)
Measure of a player's overall skill using statistical indicators.

Sosa 66.62
Alou 60.27
Lofton 56.76 (Pitt+Chc)
Patterson 55.09
Karros 48.79
A. Ramirez 48.01 (Pitt+Chc)
Grudzielanek 47.62
Simon 40.71 (Pitt+Chc)
Choi 39.10
Alex S. Gonzalez 32.67
R. Martinez 31.65
Goodwin 27.41
D. Miller 23.31
O'Leary 18.93
Bako 16.86

Total Average (TA)
Thomas Boswell statistic that is a ratio of Total Bases accumulated divided by the outs created.

Sosa .932
Patterson .858
Lofton .832 (Pitt+Chc)
Alou .803
Choi .777
Karros .745
Goodwin .744
Ramirez .736 (Pitt+Chc)
Grudzielanek .713
Simon .653 (Pitt+Chc)
Alex S. Gonzalez .647
R. Martinez .636
D. Miller .630
O'Leary .581
Bako .576

Run Element Ratio (RER)
Bill James' statistic that calculates whether a player's secondary skills are more valauble early or late in an inning. The higher the ratio, the more valuable the player will early in an inning and players are 1.00 are equally valuable early or late in an inning.

Goodwin 2.31
Bako 1.16
Choi .927
Lofton .905 (Pitt+Chc)
Martinez .889
Miller .833
Grudzielanek .735
O'Leary .708
Alou .641
Karros .537
Alex S. Gonzalez .515
Patterson .443
Sosa .437
A. Ramirez .376 (Pitt+Chc)
Simon .246 (Pitt+Chc)

Linear Weights (LWTS)
Palmer and Thorn method of evaulating offensive production with weights given to various events and then adding them up.

Sosa 93.88
Alou 86.21
A. Ramirez 82.34 (Pitt+Chc)
Lofton 81.44 (Pitt+Chc)
Grudzielanek 63.26
Alex S. Gonzalez 60.71
Patterson 50.65
Simon 48.38 (Pitt+Chc)
Karros 46.38
Miller 39.37
R. Martinez 33.32
Choi 29.65
Goodwin 21.01
Bako 18.04
O'Leary 17.16

Linear Weights per 100 Plate Appearences
Since LWTS is dependent on the # of Plate Appearences, I decided to make it per 100PAs to get a better idea of who was more productive if they had the same # of PAs.

Sosa 15.93
Patterson 14.59
Alou 13.50
Lofton 13.34 (Pitt+Chc)
Karros 12.70
A. Ramirez 12.28 (Pitt+Chc)
Choi 12.10
Grudzielanek 11.91
Goodwin 11.41
Simon 11.22 (Pitt+Chc)
Alex S. Gonzalez 10.09
R. Martinez 10.00
D. Miller 9.84
O'Leary 8.84
Bako 8.46

Runs Created (RC)
Bill James stat that estimates the # of runs a player would create given his offensive stats

Sosa 94.5
Alou 91.7
Lofton 90.5 (Pitt+Chc)
A. Ramirez 87.2 (Pitt+Chc)
Grudzielanek 72.3
Alex S. Gonzalez 61.7
Patterson 54.3
Simon 53.5 (Pitt+Chc)
Karros 46.9
D. Miller 37.9
R. Martinez 36.8
Choi 31.8
Goodwin 21.9
Bako 20
O'Leary 15.9

Runs Created per Game (RC/G)
An idea of how how important a particular player is to a lineup. For example, Hee Seop Choi has a RC/G of 5.26 and if you to take 9 Hee Seop Chois and put them in the lineup, that team would likely score around 5.26 runs per game.

Sosa 6.74
Patterson 6.03
Lofton 5.91 (Pitt+Chc)
Alou 5.8
Grudzielanek 5.53
Choi 5.26
A. Ramirez 4.95 (Pitt+Chc)
Karros 4.95
Simon 4.74 (Pitt+Chc)
Goodwin 4.48
R. Martinez 4.27
Alex S. Gonzalez 3.73
Bako 3.59
D. Miller 3.49
O'Leary 2.89

Offensive Winning % (OWP)
The % of games a team would win with 9 of those players in the lineup with average pitching and defense (based from RC/G). This happens to be one of my fav. stats for mostly entertainment purposes.

Sosa .681 (110-52)
Patterson .631 (102-60)
Lofton .622 (100-62) (Pitt+Chc)
Alou .622 (99-63)
Grudzielanek .590 (95-67)
Choi .566 (91-71)
A. Ramirez .536 (86-76) (Pitt+Chc)
Karros .536 (86-76)
Simon .514 (83-79) (Pitt+Chc)
Goodwin .486 (78-84)
R. Martinez .462 (74-88)
Alex S. Gonzalez .396 (64-98)
Bako .378 (61-101)
Miller .364 (58-104)
O'Leary .282 (45-117)

Adjusted OPS (aOPS)
This is a statistic I created that is basically designed to add weight to OBP (1.75) into OPS and the extra step of putting it in a format similar as to why Equiv. Avg. (EqA) is in a batting avg. format.

Sosa .306
Patterson .282
Alou .282
Lofton .277 (Pitt+Chc)
Grudzielanek .274
Karros .270
Choi .268
A. Ramirez .268 (Pitt+Chc)
Simon .253 (Pitt+Chc)
R. Martinez .249
Goodwin .243
Alex S. Gonzalez .240
D. Miller .237
Bako .227
O'Leary .217

(Editor's Note: I only used 3 weighted stats (BR, LWTS, and RC), I felt given the close proximity of most weighted stats that from these 3 the reader would be able to differentiate the production of players based on weighted figures. This is the reason why I did not use XR/XRB/XRR by Jim Furtado and Baseruns by David Smyth (which Baseruns is likely the most effective when used with a dominant seasons like Bonds of 02'.)


1. Hendry has given every indication that the bullpen is the #1 priority of the off-season, the offense has to become at worst an equal priority as the bullpen for the Cubs to improve in 04'. Based on my calculations, with this offense if you had given this offense avg. pitching and defense, the Cubs would have finished with 78-79 wins (which is only 3-4 wins better than the pythag. record of 02'). It is quite obvious improving the ability to get on base would benefit this team more than any other adjustment possible.

2. Sosa is still the most prolific offensive weapon on the Cubs. He does it while in the declining stages of his career, his bat speed has declined to the point where he is resided to cheating (insert cork jokes) as far as getting the bat head in motion early to catch up to a pitcher w/a high velocity FB. He is good, but not great anymore and good enough to continue to be the Cubs' best hitter.

3. This offense would likely be good enough to become a strength with the top 6 hitters (Sosa, Patterson, Alou, Grudzielanek, Ramirez, and Choi) returning from last year's line-up. While it is difficult to project the projection levels of Patterson, Choi and Grudzielanek next year, it can be satisfactory if they decide to-sign Grudzielanek again and give Patterson/Choi the bulk of ABs at their respective positions they can be good enough not to become a weakness.

4. While the offense is good enough for the 6 previous mentioned, it is not good enough to have the dead weight (offense) of having both Miller and Gonzalez at the bottom of the order. One has to be replaced, a 7-8-9 of Gonzalez, Miller, and a SP will not get it done and while the Cubs have an opening at 2B, there would not be much improvement of any of the FA possibilities to replace the production that Grudzielanek provided last year. A substantial improvement at either C or SS is key to next year's success.

5. Choi is fine at 1B. I cannot stress this enough at how important it would be to find out if Choi can handle the position. Overall, Choi had a successful 1st season and until Baker decided to switch how he was going to use Choi (after Choi's injury) did Choi start to struggle. At the beginning of the year Choi was used against most RH pitchers as Karros was used against most LHs and the few RHs he had success against. After Choi's injury, Baker for some reason that boggles my mind decided to use Karros against the LHs and instead of using Karros only against RHs he has hit well in the past, he uses Karros against every RH that he hasn't struggled against. Now, isn't it obvious that Karros would likely struggle and have lower offensive numbers vs. above avg. RH starters compared to mediocre or poor RH starters? This is setting Choi up to fail and it is exactly what it did, looking at the Game Log for Choi in the second half of the season it is obvious that Baker used him mostly against the better RH pitchers of the NL, which is impossible for anyone not named Barry Bonds to succeed against. (Thanks to Tim @ NorthsideBaseball.com who helped provide the Game Log information.)

6. Help wanted: Power off the bench.

This page is powered by Blogger. Isn't yours?

Site Meter