The Stats That Matter Most

Many moons ago, trendsetting sabermetrician Bill James demonstrated a relationship between runs scored, runs allowed and wins which, because of the necessity of squaring the elements, he dubbed the Pythagorean Theorem (of baseball). The formula is that the square of runs scored divided by the sum of the square of runs scored and the square of runs allowed equals a team's expected winning percentage. From there, you simply multiply the winning percentage and games played and you have a projected win total. The measure works remarkably well, and, obviously, the more games a team plays, the better the relationship ordinarily is.

AMERICAN LEAGUE
Team	R	OR	R DIFF	G	R/G	OR/G	DIFF/G	PW%	EXP W	ACT W	W DIFF
Anaheim Angels	836	734	102	162	5.16	4.53	0.63	.565	91	92	1
Baltimore Orioles	842	830	12	162	5.20	5.12	0.07	.507	82	78	-4
Boston Red Sox	949	768	181	162	5.86	4.74	1.12	.604	98	98	0
Chicago White Sox	865	831	34	162	5.34	5.13	0.21	.520	84	83	-1
Cleveland Indians	858	857	1	162	5.30	5.29	0.01	.501	81	80	-1
Detroit Tigers	827	844	-17	162	5.10	5.21	-0.10	.490	79	72	-7
Kansas City Royals	720	905	-185	162	4.44	5.59	-1.14	.388	63	58	-5
Minnesota Twins	780	715	65	162	4.81	4.41	0.40	.543	88	92	4
New York Yankees	897	808	89	162	5.54	4.99	0.55	.552	89	101	12
Oakland Athletics	793	742	51	162	4.90	4.58	0.31	.533	86	91	5
Seattle Mariners	698	823	-125	162	4.31	5.08	-0.77	.418	68	63	-5
Tampa Bay Devil Rays	714	842	-128	161	4.43	5.23	-0.80	.418	67	70	3
Texas Rangers	860	794	66	162	5.31	4.90	0.41	.540	87	89	2
Toronto Blue Jays	719	823	-104	161	4.47	5.11	-0.65	.433	70	67	-3
	11358	11316	42	1133	5.01	4.99

NATIONAL LEAGUE
Team	R	OR	R DIFF	G	R/G	OR/G	DIFF/G	PW%	EXP W	ACT W	W DIFF
Arizona Diamondbacks	615	899	-284	162	3.80	5.55	-1.75	.319	52	51	-1
Atlanta Braves	803	668	135	162	4.96	4.12	0.83	.591	96	96	0
Chicago Cubs	789	665	124	162	4.87	4.10	0.77	.585	95	89	-6
Cincinnati Reds	750	907	-157	162	4.63	5.60	-0.97	.406	66	76	10
Colorado Rockies	833	923	-90	162	5.14	5.70	-0.56	.449	73	68	-5
Florida Marlins	718	700	18	162	4.43	4.32	0.11	.513	83	83	0
Houston Astros	803	698	105	162	4.96	4.31	0.65	.570	92	92	0
Los Angeles Dodgers	761	684	77	162	4.70	4.22	0.48	.553	90	93	3
Milwaukee Brewers	634	757	-123	161	3.94	4.70	-0.76	.412	66	67	1
Montreal Expos	635	769	-134	162	3.92	4.75	-0.83	.405	66	67	1
New York Mets	684	731	-47	162	4.22	4.51	-0.29	.467	76	71	-5
Philadelphia Phillies	840	781	59	162	5.19	4.82	0.36	.536	87	86	-1
Pittsburgh Pirates	680	744	-64	161	4.22	4.62	-0.40	.455	73	72	-1
San Diego Padres	768	705	63	162	4.74	4.35	0.39	.543	88	87	-1
San Francisco Giants	850	770	80	162	5.25	4.75	0.49	.549	89	91	2
St. Louis Cardinals	855	659	196	162	5.28	4.07	1.21	.627	102	105	3
	12018	12060	-42	1295	4.64	4.66

Orioles Comments: The Orioles finished the season with their most disparate WDIFF of the year (-4). The Pythagorean Theorem projected 82 wins for the O's, four more than the 78 they actually amassed.

The Orioles' 5.2 R/G mark was modestly better than the league average (5.01) and ranked the team sixth in the American League. That mark was the team's high water level of the year based on the previous Stats that Matter Most reports. The Birds had been eighth in the league in runs from late June on, so they ticked up a bit in that category late in the season.

In terms of runs allowed, the Orioles finished with a rank of nine in the AL, a truly remarkable improvement in comparison with the club's status earlier in the season. To review, on the morning of June 25 (a date corresponding almost identically with the replacement of pitching coach Mark Wiley with Ray Miller), the Orioles were dead last in the league in OR/G with a mark of 5.87. By early August, the rate was down to 5.47 and by early September it had dropped to 5.22. The final mark was 5.12.

From June 25 through the end of the season, the Orioles performed markedly better than they had from the season opener through June 24. Here's the Orioles' line for these two segments of the season:

The offense, as you can see, improved slightly, but the real step forward came in the OR/G component. Over the final 94 games, the Orioles lowered their runs allowed by 1.28 per contest. 4.59 OR/G, projected over the entire season, would have ranked the Orioles fourth in the league, .001 behind the third place Oakland A's. The Birds' performance over their final 94 games, extended over a 162-game schedule, shows a Pythagorean projection of 92 wins. Only one team in the AL had a projected win total of 92+ games in 2004 (Boston). In other words, over those final 94 games, the Orioles played at a playoff caliber level. Food for thought.

AL Notable Numbers: Incredibly, the Yankees' WDIFF actually increased after early September. New York's final WDIFF mark of +12 is one of the highest you'll ever see. It's not that the Yankees were a bad club...they just weren't a particularly good one, and they performed at nothing close to the 100-win level that they surpassed this season. New York finished right at the league average in OR/G, and won on the strength of the club's second ranked offense. But the Yankees' performance was so far below that of the team they beat out (the Red Sox) as to be truly incredible. Boston's run differential was 92 runs larger than New York's.

Detroit was at the other end of the spectrum, finishing -7 in WDIFF. The Tigers were, legitimately, pretty close to a .500 team in 2004, a remarkable improvement over 2003 when they were--again, legitimately--pretty close to a 120-loss club. Kansas City and Seattle both finished with -5 WDIFF marks, but both teams were truly lousy...just not quite as lousy as their final W-L records.

The team with the blaring warning bells--along with the Yankees of course (there's just no way to ignore a +12 WDIFF)--is the Oakland A's who missed the playoffs this year despite the benefit of a +5 WDIFF. The A's appear to finally be feeling the effects of losing productive players in recent years.

NL Notable Numbers: The Cincinnati Reds are the NL version of the Yankees, less the 100 real life wins. When you lose 86 games despite a WDIFF of +10, you've got major, major problems. In the National League, only pitiful Arizona (a legitimate 110-loss club) had a worse RDIFF than the Reds who allowed almost as many runs as the Colorado Rockies without the excuse of playing in Coors Field.

The Rockies and the New York Mets both finished with -5 WDIFF marks but since neither projected to even a .500 record with the correction, it's difficult to get particularly excited.

The Chicago Cubs, on the other hand, finished with a -6 WDIFF, projecting to 95 wins (a total that would have been enough to net them a postseason slot had they been able to convert that projection into actual on-the-field wins).

Most of the other clubs in both leagues finished with win totals within a few games of their Pythagorean projections, so their final records can be seen as essentially commensurate with their run scoring/allowing totals.

	R	OR	RDIFF	G	R/G	OR/G	DIFF/G	PW%	EXP W	ACT W	W DIFF
Thru 6/24	349	399	-50	68	5.13	5.87	-0.74	.433	29	29	0
From 6/25	493	431	+62	94	5.24	4.59	+0.66	.567	53	49	-4