How Often Should Your Team Win The World Series?

The World Series is special. Some teams are overdue, but not as many as you’d think. (via M&R Glasgow)

At the end of the 2015 World Series, the Kansas City Royals and their fans celebrated a long-awaited victory. It was the sort of triumph dreamed of by all stalwart fans of losing clubs, a world championship won, at long last, with a talented group of homegrown young players after decades of futility. Headlines across the nation celebrated the end of the Royals’ 30-year drought.

But something about that “30-year drought” catches the eye. There are 30 teams in major league baseball. If every team had an equal chance of winning, could 30 years really be called a drought? If you look at it that way, the Royals seemed to be hoisting the Commissioner’s Trophy precisely on schedule. The Cubs’ 108 years without a World Series win? Now, that’s a drought.

How many titles should the average team win? This called for a deeper look at the data. Because the number of teams has fluctuated since the first World Series in 1903, it is not as simple as just dividing the number of seasons by 30 and seeing which team is on target. Before expansion, the average team had a chance of winning the World Series once every 16 years. The odds lengthened to one-in-18, one-in-20, and so on until the majors arrived at their present size in 1998.

Expected Wins by Years in MLB

TEAM	03-’60	61	62-’68	69-’76	77-’92	93-’97	98-’15	TOTAL	EXP.	EXP. ROUND	PERF.	OVER/ UNDER	DIFF.
Yankees	18	1	1	0	2	1	4	27	5.7	6	477.1%	21	21
Cardinals	6	0	2	0	1	0	2	11	5.7	6	194.4%	5	5
Athletics	5	0	0	3	1	0	0	9	5.7	6	159.0%	3	3
Red Sox	5	0	0	0	0	0	3	8	5.7	6	141.4%	2	2
Giants	5	0	0	0	0	0	3	8	5.7	6	141.4%	2	2
Marlins	-	–	–	–	-	1	1	2	0.7	1	269.2%	1	1
Blue Jays	-	–	–	–	1	1	0	2	1.4	1	147.3%	1	1
Dbacks	-	–	–	–	-	–	1	1	0.6	1	166.7%	0	0
Royals	-	–	–	0	1	0	1	2	1.7	2	118.2%	0	0
Dodgers	2	0	2	0	2	0	0	6	5.7	6	106.0%	0	0
Mets	-	–	0	1	1	0	0	2	2.0	2	97.9%	0	0
Reds	2	0	0	2	1	0	0	5	5.7	6	88.3%	-1	1
Pirates	3	0	0	1	1	0	0	5	5.7	6	88.3%	-1	1
Angels	-	0	0	0	0	0	1	1	2.1	2	47.7%	-1	1
Mariners	-	–	–	–	0	0	0	0	1.4	1	0.0%	-1	1
Rockies	-	–	–	–	-	0	0	0	0.7	1	0.0%	-1	1
Rays	-	–	–	–	-	–	0	0	0.6	1	0.0%	-1	1
Tigers	2	0	1	0	1	0	0	4	5.7	6	70.7%	-2	2
Rangers*	-	0	0	0	0	0	0	0	2.1	2	0.0%	-2	2
Astros	-	–	0	0	0	0	0	0	2.0	2	0.0%	-2	2
Nationals^	-	–	–	0	0	0	0	0	1.7	2	0.0%	-2	2
Brewers	-	–	–	0	0	0	0	0	1.7	2	0.0%	-2	2
Padres	-	–	–	0	0	0	0	0	1.7	2	0.0%	-2	2
Orioles”	0	0	1	1	1	0	0	3	5.7	6	53.0%	-3	3
White Sox	2	0	0	0	0	0	1	3	5.7	6	53.0%	-3	3
Twins**	1	0	0	0	2	0	0	3	5.7	6	53.0%	-3	3
Braves	2	0	0	0	0	1	0	3	5.7	6	53.0%	-3	3
Indians	2	0	0	0	0	0	0	2	5.7	6	35.3%	-4	4
Phillies	0	0	0	0	1	0	1	2	5.7	6	35.3%	-4	4
Cubs	2	0	0	0	0	0	0	2	5.7	6	35.3%	-4	4
TOTAL	28	0	5	8	13	3	14	71	105.3	112

* = Senators/Rangers
^ = Expos/Nationals
” = Browns/Orioles
** = Senators/Twins

As expected, the Royals were right on target. They joined the American League in 1969, one of 24 major league teams at the time. Since then, each team could be expected to average 1.692 World Series victories. With two, the Royals were slightly ahead of the average. Their 2015 National League opponents, the Mets, were also just off the average, despite their loss that year, with two World Series wins against an expected 2.042.

A few other figures jump out at you from that chart. The Yankees, unsurprisingly, had far more World Series wins than an average team could expect. With 27 championships, the only way the Yankees would fall below their expected number of World Series wins is if they don’t win their 28th until the year 2656. Their legacy is secure, for our lifetimes at least.

The other above-average teams include some you would expect, given their long history of winning baseball: the Cardinals, Athletics, Giants and Red Sox. The other three overachievers are teams that won one of two World Series victories in an unusually short time since they joined the league: the Marlins, Blue Jays and Diamondbacks. Besides the Royals and Mets, only the Dodgers are roughly on target with six wins against an expected 5.66. The teams farthest back are also unsurprising: the Phillies, Cubs and Indians have all won two world championships in a span of 111 World Series, when they should have been expected to win 5.66.

This shows us something about which teams have dominated their time in baseball, but it also shows us something else: the number of years in the majors does not accurately predict how many times a team has won the World Series. The average team’s predicted World Series total was off by 2.567 World Series, making the measurement interesting historically, but without predictive value. If all teams are created equal, yes, then every team has one chance in 30 of winning, but as baseball fans know, all teams are far from equal. Maybe, just maybe, the team with the best players is most likely to win.

I ran those numbers, with the team leading in team WAR in a given year standing in for a judgment of which team had the “best players.” Given the premium paid to pitchers lately, one might be forgiven for thinking that the team with the best pitching staffs over the years would have the most world championships. You would be wrong.

Expected Wins by Pitching WAR

TEAM	LED IN WAR (pre-expansion)	LED IN WAR (post-expansion)	EXPECTED WS (based on WAR)	ACTUAL WS	PERFORMANCE	OVER/ UNDER	DIFF.
Yankees	2	6	8	27	337.50%	19	19
Cardinals	4	1	5	11	220.00%	6	6
Athletics	5	0	5	9	180.00%	4	4
Browns/Orioles	0	0	0	3	INF	3	3
Dodgers	0	3	3	6	200.00%	3	3
Giants	6	0	6	8	133.33%	2	2
Marlins	-	0	0	2	INF	2	2
Diamondbacks	-	0	0	1	INF	1	1
Red Sox	8	0	8	8	100.00%	0	0
Royals	-	2	2	2	100.00%	0	0
Brewers	-	0	0	0	100.00%	0	0
Blue Jays	-	2	2	2	100.00%	0	0
Mariners	-	0	0	0	100.00%	0	0
Rockies	-	0	0	0	100.00%	0	0
Rays	-	0	0	0	100.00%	0	0
Reds	4	2	6	5	83.33%	-1	1
Pirates	6	0	6	5	83.33%	-1	1
Angels	-	2	2	1	50.00%	-1	1
Senators/Rangers	-	1	1	0	0.00%	-1	1
Astros	-	1	1	0	0.00%	-1	1
Mets	-	3	3	2	66.67%	-1	1
Expos/Nationals	-	1	1	0	0.00%	-1	1
Padres	-	1	1	0	0.00%	-1	1
Senators/Twins	1	4	5	3	60.00%	-2	2
Phillies	2	2	4	2	50.00%	-2	2
White Sox	3	5	8	3	37.50%	-5	5
Tigers	6	3	9	4	44.44%	-5	5
Indians	7	1	8	2	25.00%	-6	6
Braves	0	9	9	3	33.33%	-6	6
Cubs	3	5	8	2	25.00%	-6	6
TOTAL	57	54	111	111			2.667

Forget what you’ve heard: Pitching does not win championships. Or at least it doesn’t predict them. There are more teams precisely on target (seven,) but the average team’s predicted World Series win total is off by more than the previous analysis. The Royals, again, won two championships, just as predicted, but most of those on target are teams that both never had the best pitching and never won the World Series (the Brewers, Mariners, Rockies, and Rays). The teams predicted to win the most Series by this measure, the Tigers and the Braves, had top pitching nine times each, but won only four and three times, respectively.

Maybe offense is the answer. Substituting batters’ WAR in place of pitchers’, we get a much more predictive result:

Expected Wins by Batting WAR

TEAM	LED IN WAR (pre-expansion)	LED IN WAR (post-expansion)	EXPECTED WS (based on WAR)	ACTUAL WS	PERFORMANCE	OVER/ UNDER	DIFF.
Cardinals	2	3	5	11	220.00%	6	6
White Sox	0	0	0	3	INF	3	3
Yankees	24	1	25	27	108.00%	2	2
Senators/Twins	0	1	1	3	300.00%	2	2
Athletics	5	2	7	9	128.57%	2	2
Dodgers	3	1	4	6	150.00%	2	2
Royals	-	0	0	2	INF	2	2
Marlins	-	0	0	2	INF	2	2
Red Sox	2	5	7	8	114.29%	1	1
Blue Jays	-	1	1	2	200.00%	1	1
Diamondbacks	-	0	0	1	INF	1	1
Phillies	0	2	2	2	100.00%	0	0
Cubs	2	0	2	2	100.00%	0	0
Pirates	3	2	5	5	100.00%	0	0
Angels	-	1	1	1	100.00%	0	0
Senators/Rangers	-	0	0	0	100.00%	0	0
Astros	-	0	0	0	100.00%	0	0
Expos/Nationals	-	0	0	0	100.00%	0	0
Padres	-	0	0	0	100.00%	0	0
Rockies	-	0	0	0	100.00%	0	0
Tigers	1	4	5	4	80.00%	-1	1
Mets	-	3	3	2	66.67%	-1	1
Rays	-	1	1	0	0.00%	-1	1
Browns/Orioles	0	5	5	3	60.00%	-2	2
Braves	2	3	5	3	60.00%	-2	2
Indians	3	2	5	2	40.00%	-3	3
Reds	2	6	8	5	62.50%	-3	3
Brewers	-	3	3	0	0.00%	-3	3
Mariners	-	3	3	0	0.00%	-3	3
Giants	8	5	13	8	61.54%	-5	5
TOTAL	57	54	111	111			1.6

Instead of the 5.66 wins predicted by longevity or the eight predicted by pitching WAR, this model shows that the Yankees led the league in batting WAR 25 times, which is close to the 27 championships they actually won. Nine teams won exactly the number predicted, more than in the previous comparisons. Among the biggest misses are the Giants, who had the league’s best hitting 13 times but won only five World Series, and the Mariners, who led the majors in batting WAR in 1996, 1997 and 2001, but failed to bring home a single world championship. On the other side, the White Sox, Royals, Marlins and Diamondbacks all managed to come out on top in the postseason at least once despite never having the best team according to batting WAR.

Even more interesting is that the years a team had the best WAR and the years it won the World Series do not always match up. The Yankees’ predicted number is close to their actual total, but 24 of their 25 WAR-leading seasons came before the expansion era began. Since that time, the Bronx Bombers have won nine world championships, but led the league in batting WAR only once. Similarly, the Phillies’ two WAR-leading years are equal to their number of World Series rings, but the batting WAR titles came in different years (1977 and 2007) than their World Series titles (1980 and 2008).

The Expos/Nationals met their batting WAR-predicted totals exactly: zero years with the best WAR, and zero world championship titles. Many of the remaining Montreal baseball fans see 1994, the strike-shortened year, as the beginning of the end for the Expos, who had the best record in baseball when the season was cancelled. The Expos ranked sixth in WAR that year. The team at the top of the WAR standings was the Houston Astros, the only time in team history that they led in that category. Astros fans may also have a reason to curse that annus horribilis.

The batters’ WAR measurement predicted World Series wins better than the other two, but was still off by an average of 1.6 Series wins per team. If the aggregate performance of individual players does not give us the answer, maybe we should look to an old-fashioned statistic. Does having the most wins in the majors predict victory in the World Series? No, but it gets us closer.

Expected Wins by Wins

TEAM	LED IN W/L (pre-expansion)	LED IN W/L (post-expansion)	EXPECTED WS (based on W/L)	ACTUAL WS	PERFORMANCE	OVER/ UNDER	DIFF.
Red Sox	4	1	5	8	160.00%	3	3
Giants	3	2	5	8	160.00%	3	3
Cardinals	3	5.5	8.5	11	129.41%	2.5	2.5
Pirates	1	2	3	5	166.67%	2	2
Blue Jays	-	0	0	2	INF	2	2
Marlins	-	0	0	2	INF	2	2
Dodgers	4	1	5	6	120.00%	1	1
Royals	-	1	1	2	200.00%	1	1
Diamondbacks	-	0	0	1	INF	1	1
Mets	-	1.5	1.5	2	133.33%	0.5	0.5
White Sox	2	1	3	3	100.00%	0	0
Tigers	1	3	4	4	100.00%	0	0
Senators/Twins	2	1	3	3	100.00%	0	0
Athletics	6	3	9	9	100.00%	0	0
Phillies	0	2	2	2	100.00%	0	0
Senators/Rangers	-	0	0	0	100.00%	0	0
Astros	-	0	0	0	100.00%	0	0
Padres	-	0	0	0	100.00%	0	0
Rockies	-	0	0	0	100.00%	0	0
Rays	-	0	0	0	100.00%	0	0
Yankees	19	9	28	27	96.43%	-1	1
Reds	2	4	6	5	83.33%	-1	1
Angels	-	2	2	1	50.00%	-1	1
Expos/Nationals	-	1	1	0	0.00%	-1	1
Brewers	-	1	1	0	0.00%	-1	1
Mariners	-	1	1	0	0.00%	-1	1
Browns/Orioles	0	5	5	3	60.00%	-2	2
Braves	1	4.5	5.5	3	54.55%	-2.5	2.5
Indians	3	2.5	5.5	2	36.36%	-3.5	3.5
Cubs	6	0	6	2	33.33%	-4	4
TOTAL	26	54	111	111			1.2

This method is a lot more predictive than any of the previous attempts (where two teams tied for the WAR lead, they are each credited with .5). Nine teams won exactly the number of World Series their win totals suggest they should, the same as in the batting WAR example, but this time the average discrepancy is even smaller: 1.2, compared to 1.6. The Yankees’ expected total is almost exactly equal to the titles they actually won: 28 instead of 27. The biggest miss with this prediction method is the Indians, who led the majors in wins five times and tied for the lead once, but won only two world championships.

That having the most wins in a season predicts World Series victories is not surprising. Before 1969, having the top record guaranteed at least a World Series appearance, leading to a roughly 50/50 shot of winning it all. Since that was no longer true once additional rounds of the playoffs were added, you would think that wins would be less predictive in after 1968, when the team with the best record was no longer guaranteed a spot in the Series. Strangely, this is not true.

Expected Wins by Wins since 1961

TEAM	LED IN W/L (post-expansion)	ACTUAL WS	PERFORMANCE	OVER/ UNDER	DIFF.
Red Sox	1	3	300.00%	2	2
Senators/Twins	0	2	INF	2	2
Giants	1	3	300.00%	2	2
Blue Jays	0	2	INF	2	2
Marlins	0	2	INF	2	2
Yankees	6	7	116.67%	1	1
Athletics	3	4	133.33%	1	1
Dodgers	1	2	200.00%	1	1
Royals	1	2	200.00%	1	1
Diamondbacks	0	1	INF	1	1
Mets	1.5	2	133.33%	0.5	0.5
White Sox	1	1	100.00%	0	0
Phillies	2	2	100.00%	0	0
Cubs	0	0	100.00%	0	0
Pirates	2	2	100.00%	0	0
Senators/Rangers	0	0	100.00%	0	0
Astros	0	0	100.00%	0	0
Padres	0	0	100.00%	0	0
Rockies	0	0	100.00%	0	0
Rays	0	0	100.00%	0	0
Tigers	2	1	50.00%	-1	1
Reds	4	3	75.00%	-1	1
Angels	2	1	50.00%	-1	1
Expos/Nationals	1	0	0.00%	-1	1
Brewers	1	0	0.00%	-1	1
Mariners	1	0	0.00%	-1	1
Cardinals	4.5	3	66.67%	-1.5	1.5
Browns/Orioles	4	2	50.00%	-2	2
Indians	2.5	0	0.00%	-2.5	2.5
Braves	4.5	1	22.22%	-3.5	3.5
TOTAL	46	46			1.033

As before, nine teams are exactly on target in World Series victories since 1969, and the average variation is 1.033 Series wins. The variation is likely smaller in part because fewer years being measured gives less time for inaccuracies to accumulate, but the largest variation is the same as when measuring since 1903: since 1969 the Braves have underachieved the most, leading the majors in wins four times and tying for the lead once, but have won just one World Series, in 1995.

If you are familiar with advanced statistics, you may now be thinking that having the most wins is not necessarily indicative of having the best team. The concept of Pythagorean Wins is one attempt to fix this problem. Pythagorean Wins (explained more fully here) measures the runs a team scores and the runs it surrenders and gives us a rough tally of how many wins and losses the team “should have had” based on those figures.

Expected Wins by Pythagorean Wins

TEAM	LED IN PythWL (pre-expansion)	LED IN PythWL (post-expansion)	EXPECTED WS (based on PythWL)	ACTUAL WS	PERFORMANCE	OVER/ UNDER	DIFF.
Red Sox	3	1	4	8	200.00%	4	4
Yankees	17.5	6	23.5	27	114.89%	3.5	3.5
Giants	3	2	5	8	160.00%	3	3
Cardinals	3.5	5	8.5	11	129.41%	2.5	2.5
Reds	2	1	3	5	166.67%	2	2
Royals	-	0	0	2	INF	2	2
Marlins	-	0	0	2	INF	2	2
Tigers	1	2	3	4	133.33%	1	1
Pirates	2	2	4	5	125.00%	1	1
Senators/Twins	2	1	3	3	100.00%	0	0
Athletics	6	3	9	9	100.00%	0	0
Braves	1	2	3	3	100.00%	0	0
Phillies	0	2	2	2	100.00%	0	0
Angels	-	1	1	1	100.00%	0	0
Senators/Rangers	-	0	0	0	100.00%	0	0
Astros	-	0	0	0	100.00%	0	0
Mets	-	2	2	2	100.00%	0	0
Padres	-	0	0	0	100.00%	0	0
Blue Jays	-	2	2	2	100.00%	0	0
Rockies	-	0	0	0	100.00%	0	0
Diamondbacks	-	1	1	1	100.00%	0	0
Rays	-	0	0	0	100.00%	0	0
White Sox	2	1.5	3.5	3	85.71%	-0.5	0.5
Expos/Nationals	-	1	1	0	0.00%	-1	1
Dodgers	4	4	8	6	75.00%	-2	2
Brewers	-	2	2	0	0.00%	-2	2
Mariners	-	2	2	0	0.00%	-2	2
Indians	3	2	5	2	40.00%	-3	3
Cubs	6	1	7	2	28.57%	-5	5
Browns/Orioles	1	7.5	8.5	3	35.29%	-5.5	5.5
TOTAL	41	54	111	111			1.4

Pythagorean wins correctly predicted the number of World Series wins for 13 teams, the best of any method tried. The average error (1.4 Series wins) was higher, however, than the rate when using just actual wins. That may be also attributed to the pre-1969 guarantee that a team with the most wins would at least earn a World Series appearance: being the “best team” didn’t matter if you finished second in your league. Running the numbers for just the expanded playoffs period narrows the gap to an average error of 1.1, almost the same as the 1.033 found when measuring actual wins. The Royals, the team that provoked this inquiry in the first place, exceeded the World Series titles predicted by Pythagorean wins, as they have never led the majors in that statistic.

Expected Wins by Pythagorean Wins since 1961

TEAM	LED IN PythWL (post-expansion)	ACTUAL WS	PERFORMANCE	OVER/ UNDER	DIFF.
Yankees	4	7	175.00%	3	3
Red Sox	1	3	300.00%	2	2
Senators/Twins	0	2	INF	2	2
Reds	1	3	300.00%	2	2
Giants	1	3	300.00%	2	2
Royals	0	2	INF	2	2
Marlins	0	2	INF	2	2
Athletics	3	4	133.33%	1	1
White Sox	0.5	1	200.00%	0.5	0.5
Tigers	1	1	100.00%	0	0
Phillies	2	2	100.00%	0	0
Pirates	2	2	100.00%	0	0
Angels	1	1	100.00%	0	0
Senators/Rangers	0	0	100.00%	0	0
Astros	0	0	100.00%	0	0
Mets	2	2	100.00%	0	0
Padres	0	0	100.00%	0	0
Blue Jays	2	2	100.00%	0	0
Rockies	0	0	100.00%	0	0
Diamondbacks	1	1	100.00%	0	0
Rays	0	0	100.00%	0	0
Braves	2	1	50.00%	-1	1
Cubs	1	0	0.00%	-1	1
Cardinals	4	3	75.00%	-1	1
Expos/Nationals	1	0	0.00%	-1	1
Indians	2	0	0.00%	-2	2
Dodgers	4	2	50.00%	-2	2
Brewers	2	0	0.00%	-2	2
Mariners	2	0	0.00%	-2	2
Browns/Orioles	6.5	2	30.77%	-4.5	4.5
TOTAL	46	46			1.1

The long and short of all of this is that hanging around in the majors for a while is no predictor of World Series wins. This is not news to fans of teams currently in long droughts, like the Cubs, or recently emerged from them, like their cross-town rivals, the White Sox. But wins, Pythagorean wins, or cumulative WAR predict a World Series victory only about half the time at best. The vagaries and randomness of a seven-game postseason series flummox any method of prediction. That may be frustrating to fans of teams that seem historically deprived by these measures, but it also adds an element of genuine unpredictability to baseball that should keep fans glued to their seats until the last pitch of the last game of the World Series. Despite the great strides in statistical analysis in the past few years, postseason baseball is still a time when anything can happen.