Friday, July 31, 2015

Are Balanced Teams More Successful?

This was published in SABR's Baseball Research Journal several years ago.

If a team scores 10% more runs than average and allows 10% fewer runs than average, they could be said to be perfectly balanced. Do such teams win more games than teams that are less balanced? For example, if another team scored 15% more runs than average and allowed 5% fewer runs than average, they would obviously be less balanced than the first team. Does the first team win more games due to greater balance, even though they seem to have about the same level of performance as the first team?

To measure a team’s offensive performance, I divided their runs scored per game by the league average. Then that was park adjusted using the park factors from the Sean Lahman database. The 1980 Orioles, for example, scored 4.97 runs per game. That divided by the league average of 4.51 leaves 1.10. But their park factor was 99, meaning that 1% fewer runs were scored in their park than average. So the 1.10 was divided by .99 to get 1.114, which is then multiplied by 100 to get 111.4, meaning the Orioles were 11.4% better than average in scoring. I performed similar calculations for runs allowed. In that case, the Orioles got 111.77, meaning they gave up 11.77% fewer runs than average (I’m following the convention that Pete Palmer uses, so above 100 means the team was better than average at preventing runs). Let’s call the runs scored measure “OFF” for offense and the runs allowed measure “DEF” for defense

To measure balance, I found the difference between their OFF and DEF and then found the absolute value. Let’s call this result “BAL” for balance. So the closer OFF and DEF are to each other, the more balanced the team is. The Orioles had a BAL of .374 (slightly different than the numbers above would imply due to rounding). Is this balance factor important or relevant? To test this, I first ran a regression in which team winning percentage was the dependent variable and OFF and DEF were the dependent variables. The equation was:

(1) Pct = -.476 + .49*OFF + .482*DEF

(I divided both OFF and DEF by 100 for the regression so, for example, instead of using 110 for BAL, I used 1.10). The coefficient on DEF is not negative for reasons explained in above. The standard error for 162 games was about 4 wins. I looked at all teams from 1980-2004.

Then I ran the regression with the balance variable added in. Here are the results:

(2) Pct = -.476 + .486*OFF + .488*DEF - .032*BAL

The standard error was still just about 4 wins for 162 games. It did fall by about .02 wins. So adding in a balance factor does not explain winning much better. The BAL variable was statistically significant with a T-value of -2.6. It has the right sign needed if balance is to help winning. As BAL gets larger (teams get less balanced), they win less. But notice that its impact is only about 1/16 of OFF and DEF. Adding BAL also had very little impact on the equation itself, which you can see by comparing equation (2) to equation (1).

It is also helpful to look at how much a one standard deviation increase in any of the variables would change a team’s winning percentage. Standard deviation (SD) is a measure of dispersion or how spread of the numbers. The SD of OFF was 9.51. If we multiply that by .486 we get 4.62. Since I divided both OFF and DEF by 100 for the regression, we have to divide 4.62 by 100, which leaves .0462. Over a 162 games, that is about 7.5 more wins. The SD of DEF was 9.07. That multiplied by .488 leaves 4.43. Over a full season, that is 7.17 more wins. For BAL, the SD was 8.11. Using the -.032 coefficient, over 162 games we get about .42 more wins. So if you made a significant improvement in how balanced your team is you add less than 1 win a season.

I also looked to see if teams that exceeded their “Pythagorean” winning percentage were more balanced than other teams. The “Pythagorean” winning percentage was invented by Bill James and it says that a team should have a winning percentage equal to runs scored squared divided by (runs scored squared + runs squared allowed). The correlation between the BAL variable and how much teams exceeded their “Pythagorean” winning percentage was .0025, meaning that there is no connection. Being more balanced did not increase a team’s number of expected wins.

How did the most balanced teams do? The table below shows the teams with the lowest 25 BAL scores.

 Table 1: The Most Balanced Teams Rank Team Year OFF DEF BAL PCT 1 NYN 1999 110.08 110.13 0.047 0.595 2 KCA 1983 95.23 95.28 0.049 0.488 3 NYA 1997 112.67 112.62 0.050 0.593 4 SDN 1980 95.6 95.53 0.068 0.451 5 PHI 1997 89.58 89.66 0.085 0.420 6 BOS 1995 106.4 106.51 0.103 0.597 7 CHN 1987 95.21 95.32 0.113 0.472 8 TBA 2004 92.17 92 0.173 0.435 9 ATL 1985 90.49 90.27 0.224 0.407 10 TBA 1999 92.07 91.84 0.226 0.426 11 MON 1983 102.25 102.5 0.251 0.506 12 CLE 2003 95.46 95.13 0.329 0.420 13 CHN 1990 94.68 95.03 0.353 0.475 14 BAL 1980 111.4 111.77 0.374 0.617 15 NYN 1982 92.89 92.51 0.375 0.401 16 COL 1997 100.63 101.02 0.387 0.512 17 NYN 1994 97.87 98.29 0.415 0.487 18 MON 2002 100.96 101.4 0.438 0.512 19 PIT 1981 98.09 97.63 0.464 0.451 20 CAL 1996 89.65 90.14 0.486 0.435 21 PHI 1990 94.85 94.36 0.488 0.475 22 MIL 1998 94.02 93.53 0.490 0.457 23 BOS 1986 107.05 106.56 0.491 0.590 24 NYA 1982 99.76 99.26 0.502 0.488 25 CAL 1986 106.39 106.92 0.539 0.568

Their average winning percentage is .491. So they did not win any more games than normal. The next table shows the 25 least balanced teams. Their average winning percentage was .497.

Table 2: The Least Balanced Teams

 Rank Team Year OFF DEF BAL PCT 1 KCA 1987 88.32 117.16 28.845 0.512 2 DET 1993 119.10 90.18 28.924 0.525 3 TEX 1991 116.43 87.47 28.952 0.525 4 SEA 1983 73.87 103.06 29.194 0.370 5 HOU 1995 120.43 91.04 29.387 0.528 6 TEX 1983 89.22 118.80 29.585 0.475 7 TOR 1991 90.52 120.32 29.797 0.562 8 ATL 1995 94.81 124.75 29.940 0.625 9 KCA 1993 85.13 115.35 30.225 0.519 10 SLN 2003 122.14 91.04 31.096 0.525 11 CIN 2004 108.46 77.07 31.390 0.469 12 TOR 1982 82.36 113.80 31.441 0.481 13 TOR 1996 82.80 114.35 31.543 0.457 14 TEX 2001 113.04 81.34 31.702 0.451 15 BOS 1992 79.96 111.98 32.024 0.451 16 ANA 2001 82.02 114.33 32.304 0.463 17 ARI 2003 86.46 118.88 32.422 0.519 18 SFN 1999 120.86 87.80 33.068 0.531 19 BOS 1993 83.31 117.97 34.658 0.494 20 TOR 1997 81.06 116.25 35.188 0.469 21 ML4 1982 129.90 94.64 35.259 0.586 22 SDN 1997 114.63 77.00 37.635 0.469 23 TBA 1998 73.44 113.49 40.053 0.389 24 MON 2003 80.65 121.04 40.389 0.512 25 LAN 2003 82.61 126.31 43.696 0.525

The next table shows the top 25 teams in winning percentage from 1980-2004. Their average BAL score was 11.737, while the average for all teams was 10.138. So the best teams are just a little less balanced than normal (remember that zero is perfect balance).

Table 3: The Best Teams

 Rank Team Year OFF DEF BAL PCT 1 SEA 2001 126.60 118.04 8.561 0.716 2 NYA 1998 121.31 118.79 2.513 0.704 3 CLE 1995 115.25 117.67 2.416 0.694 4 NYN 1986 119.28 111.23 8.059 0.667 5 ATL 1998 108.76 128.15 19.385 0.654 6 MON 1994 109.94 116.04 6.104 0.649 7 SLN 2004 117.27 110.64 6.635 0.648 8 DET 1984 116.93 109.15 7.785 0.642 9 OAK 1988 116.84 108.16 8.674 0.642 10 ATL 1993 103.40 130.09 26.689 0.642 11 NYA 2002 118.18 108.89 9.289 0.640 12 NYA 1980 114.63 106.95 7.677 0.636 13 OAK 1990 109.59 116.13 6.540 0.636 14 SFN 1993 115.74 109.77 5.971 0.636 15 ATL 1999 110.23 116.51 6.273 0.636 16 OAK 2002 98.71 123.92 25.215 0.636 17 ATL 2002 98.83 126.79 27.959 0.631 18 HOU 1998 119.78 115.28 4.498 0.630 19 OAK 2001 116.95 118.41 1.452 0.630 20 NYN 1988 119.16 108.57 10.590 0.625 21 ATL 1995 94.81 124.75 29.940 0.625 22 SLN 1985 114.52 112.88 1.637 0.623 23 ATL 1997 103.99 128.35 24.357 0.623 24 NYA 2003 115.31 107.33 7.988 0.623 25 ATL 2003 125.16 97.93 27.225 0.623

A couple of teams are interesting here. One is the 2002-3 Braves. In 2002, they rank #17 having an OFF of just 98.83 and a DEF of 126.79. So they had great pitching and about average hitting. But the next year, 2003, ranked #25, they were very unbalanced, but in the opposite direction. They had great hitting (an OFF of 125.16) and just so-so pitching (a DEF of 97.93). The other is the 2001-02 A’s. In 2001, they were very balanced, with a BAL of 1.452. Their winning percentage was .630. The next year, they became very imbalanced when BAL rose to 25.215. But they actually saw a slight rise in their winning percentage, to .636. So for the A’s, going from being very balanced to being very imbalanced did not hurt their record.

The next table shows the lowest 25 teams in winning percentage. Their average BAL was 10.165. So the worst teams are just about as balanced as anyone else. Lack of balance is not why they lost so much.

Table 4: The Worst Teams

 Rank Team Year OFF DEF BAL PCT 1 CIN 1982 80.68 103.19 22.513 0.377 2 CLE 1987 92.56 84.60 7.962 0.377 3 SDN 1993 90.65 97.02 6.374 0.377 4 MIN 1981 78.82 99.60 20.776 0.376 5 SDN 1981 95.48 87.93 7.544 0.373 6 MIN 1982 87.96 92.09 4.132 0.370 7 SEA 1983 73.87 103.06 29.194 0.370 8 CLE 1985 99.75 85.74 14.006 0.370 9 CHN 1981 85.82 90.98 5.164 0.369 10 SEA 1980 82.23 94.47 12.234 0.364 11 DET 1989 90.49 83.56 6.930 0.364 12 NYN 1993 96.26 94.81 1.447 0.364 13 KCA 2004 93.35 86.13 7.219 0.358 14 PIT 1985 87.62 92.49 4.867 0.354 15 CLE 1991 78.49 96.68 18.193 0.352 16 TOR 1981 71.86 100.10 28.231 0.349 17 MIL 2002 90.61 85.16 5.445 0.346 18 DET 2002 79.83 84.26 4.429 0.342 19 TBA 2002 86.90 84.37 2.531 0.342 20 ATL 1988 85.11 89.68 4.567 0.338 21 BAL 1988 80.82 87.14 6.313 0.335 22 FLO 1998 93.32 78.25 15.071 0.333 23 DET 1996 90.63 79.91 10.714 0.327 24 ARI 2004 79.44 86.12 6.678 0.315 25 DET 2003 79.01 80.60 1.591 0.265

The interesting team here is the 1981-82 Twins. In 1981, their BAL was 20.776. So they were unbalanced and they had a winning percentage of just .376. The next year, their BAL fell to 4.132, meaning they became more balanced. Yet their winning percentage also fell to .370.

The most balanced team was the 1999 Mets. They boasted very good hitting and pitching, with their OFF and DEF both being just about 110. This lead to an excellent .595 winning percentage, a wild card birth in the playoffs, and a tough loss to the Braves in the NLCS (equation (1) predicts that they would have a .594 winning percentage). They boasted a star-studded lineup. The table below shows how the Met regulars hit:

 Table 5: 1999 Mets Hitting Player AB HR RBI AVG SLG OBP OPS SB Edgardo Alfonzo 628 27 108 0.304 0.502 0.385 0.886 9 John Olerud 581 19 96 0.298 0.463 0.427 0.890 3 Robin Ventura 588 32 120 0.301 0.529 0.379 0.908 1 Mike Piazza 534 40 124 0.303 0.575 0.361 0.936 2 Rey Ordonez 520 1 60 0.258 0.317 0.319 0.636 8 R. Henderson 438 12 42 0.315 0.466 0.423 0.889 37 Roger Cedeno 453 4 36 0.313 0.408 0.396 0.804 66 Brian McRae 298 8 36 0.221 0.349 0.320 0.669 2 Benny Agbayani 276 14 42 0.286 0.525 0.363 0.888 6

The Mets were a solid fifth in runs scored, averaging 5.23 runs per game.

The table below shows how the performance of the key Met pitchers:

 Table 6: 1999 Mets Pitching Pitcher W L SV IP BB SO ERA Al Leiter 13 12 0 213 93 162 4.23 Orel Hershiser 13 12 0 179 77 89 4.58 Masato Yoshii 12 8 0 174 58 105 4.40 Rick Reed 11 5 0 149 47 104 4.59 Octavio Dotel 8 3 0 85.1 49 85 5.39 Turk Wendell 5 4 3 85.1 37 77 3.07 Armando Benitez 4 3 22 78 41 128 1.85 Kenny Rogers 5 1 0 76 28 58 4.03 Pat Mahomes 8 0 0 63.1 37 51 3.71 Dennis Cook 10 5 3 63 27 68 3.86 Bobby Jones 3 3 0 59 11 31 5.64 John Franco 0 2 19 40.1 19 41 2.92

The Mets were fifth in ERA in the NL at 4.28. The pitching staff was helped by Gold Glove winners Ventura at 3B and Rey Ordonez at SS. Two other players won at least 1 Gold Glove in their careers, but not 1999, Rickey Henderson and John Olerud. The Mets also only made 68 errors that year, by far the lowest in the league, and also a record at that time. Every other team made at least 100.

The least balanced team was the 1987 Royals. They had a fairly weak hitting attack, with an OFF of just 88.32. The next table shows the Royals key hitters:

Table 7: 1987    Royals Hitting

 Player AB HR RBI AVG SLG OBA OPS SB Kevin Seitzer 641 15 83 0.323 0.470 0.399 0.869 12 Danny Tartabull 582 34 101 0.309 0.541 0.390 0.931 9 Willie Wilson 610 4 30 0.279 0.377 0.320 0.698 59 Frank White 563 17 78 0.245 0.400 0.308 0.708 1 George Brett 427 22 78 0.290 0.496 0.388 0.884 6 Bo Jackson 396 22 53 0.235 0.455 0.296 0.750 10 Steve Balboni 386 24 60 0.207 0.427 0.273 0.700 0 Jamie Quirk 296 5 33 0.236 0.345 0.307 0.652 1 Angel Salazar 317 2 21 0.205 0.246 0.219 0.465 4

Seitzer, Tartabull and Brett all had good years, but the rest of the hitters did not. The Royals were last in runs scored, averaging 4.41 runs per game (and tied for next-to-last in OPS). Their park factor was 106 that year, meaning it was slightly a better than average run environment. Now for the pitchers:

Table 8: 1987 Royals Pitching

 Pitcher W L SV IP BB SO ERA Bret Saberhagen 18 10 0 257 53 163 3.36 Mark Gubicza 13 18 0 241.2 120 166 3.99 Charlie Leibrandt 16 11 0 240.1 74 151 3.41 Danny Jackson 9 18 0 224 109 152 4.02 Bud Black 8 6 1 122.1 35 61 3.61 Steve Farr 4 3 1 91 44 88 4.15 Jerry Don Gleaton 4 4 5 50.2 28 44 4.30 Dan Quisenberry 4 1 8 49 10 17 2.76 John Davis 5 2 2 43.2 26 24 2.29 Bob Stoddard 1 3 1 40 22 23 4.28 Dave Gumpert 0 0 0 19.1 6 13 6.13 Gene Garber 0 0 8 14.1 1 3 2.55

Saberhagen, the leader of the staff, was a two-time Cy Young award winner. The Royals were second in the league in ERA at 3.87, only .13 behind the league leading Toronto Blue Jays (a little impressive since their park was favorable to hitters). Frank White (2B) won the last of his eight Gold Gloves. Willie Wilson had one career Gold Glove, but not in 1987.

(This is an expanded version of an article that originally appeared at the “Beyond the Boxscore” website)

Cyril Morong, a member of SABR since 1995, teaches economics at San Antonio College and is lifelong White Sox fan

Source:
The San Lahman database
Retrosheet
Lee Sinin’s Complete Baseball Encyclopedia
Baseball Reference