Sunday, May 27, 2007

Are the Yankees this bad or are they unlucky?

As of May 26, the Yankees are 21-26. Are they really this bad or have they been unlucky? In baseball, one can use several metrics to estimate how lucky or unlucky a team has been. Baseball Prospectus has an adjusted standings page that does a good job of updating these numbers on a daily basis. Here is the link:

They have three different adjusted standings.

The first level uses just total runs scored and total runs allowed. Using those numbers, one would expect the Yankees to go 25.9 - 21.1.

The second level uses the equivalent runs scored and allowed based on the raw statistics of each player on the Yankees and against the Yankees. In theory, this metric takes out the luck in the runs scored / runs allowed stats. Given these numbers, one would expect the Yankees to go 25.3 - 21.7.

The third level uses numbers from the second level adjusted for strength of opponents. With those adjustments, the Yankees would have gone 26.9-20.1 against an average team.

All three metrics show roughly the same thing: the Yankees were expected to win between 25.3 and 26.9 games. In fact, they have only won 21. One could interpret this as meaning that the Yankees were unlucky to the tune of between 4.3 and 5.9 wins. The Yankees look like they have been the unluckiest team this year. Two teams are close behind: the Cubs and the Reds. The Cubs have an actual record of 22-25 and an expected record of 4.5 to 5 more wins. Te Reds have an actual record of 18-32 and an expected record of 4.1 to 6.4 more wins.

Have the Yankees always been an “unlucky” team based on the statistics? This is a natural question, because if they were always unlucky, then maybe they are not truly unlucky this year. Maybe they simply do not know how to win the close ones or lack heart. However, that is not the case at all looking at the last few years. Here are the Yankees final records and expected records (using Runs Scored and Runs Allowed) from 2004 to 2006 (I only go back to 2004 because that is the earliest year that has standings with runs scored and runs allowed):

2006: Actual 97-65 Expected 96-66
2005: Actual 95-67 Expected 90-72
2004: Actual 101-61 Expected 89-73

In the last three years, one would expect the Yankees to win 18 fewer games than they actually did based on the statistics. They were lucky in those years, especially in 2004, when they were not expected to make the playoffs with only 89 expected wins. This year, they are simply getting the brunt of bad luck.

From a Yankee fan’s perspective, it is very disappointing, but the fact is that all teams have stretches or seasons where they are lucky or unlucky. The Yankees have had their share of good luck in past years by winning more games then they should have. From a fan’s perspective, it is disheartening because a fan usually only roots for one team, and the season for Yankee fans feels like it is over already. The season looks like it has slipped away and the playoffs almost out of reach. Even if they play like a 95-win team for the rest of the season, their ugly start – combined with the Red Sox hot start - will make it unlikely they get into the playoffs. In order to get into the playoffs, the Yankees will have to get as lucky (or even more so) than they have been unlucky thus far.

While I have not been reading the New York newspapers, I imagine the media is trying to come up with explanations of why this is happening. Maybe they are blaming Joe Torre and Brian Cashman. If so, did the media give them credit in 2004 for being so lucky? I doubt it. Maybe they are blaming the Yankee players for having a lack of heart. I have always hated this line of reasoning. Players want to win; they are not losing because they do not care. Players want to win more than fans want them to win.

From a gambler’s perspective, it does not seem like a big deal. Active bettors often go 20-30 in 50 game stretches. Perhaps blackjack is a better example since blackjack players play so many hands. If they go 20-30 in 50 hands, they think nothing of it. They do not go back to their hotel rooms thinking that their entire system of counting cards was lacking heart. These unlucky stretches are not that unusual, I have certainly have had plenty of those stretches in poker and sports betting. Yet, it seems like the Yankees current record, given their statistics should have an explanation. The reason it seems that way is that the sample size of a baseball season for a given team is only 162 games. A sports bettor may make 5,000 bets a year – the equivalent of 30 baseball seasons for a given team. The sports bettor will see more 20-30 stretches because he is playing many more games than the Yankees are playing. Thus, any given 20-30 stretch (while it will hurt financially and emotionally) is not very surprising.

The simple answer to the question in the title of this post is: the Yankees have been unlucky this year. While they may truly only be a 90-win team in 2007 (meaning they are not as good as they were in past years), they certainly are not as bad as they have played thus far. They are more likely to play the rest of the year like a team that should have gone 26-21 (a .553 team) rather than a team that deserves a 21-26 record (a .447 team). They have a better chance of ending close to 85 wins (the number a .447 team would have given this start) than 72 wins (the number a .447 team would have given this start). Luck is a big part of sports, and the gambler can have a higher degree of success if he can separate out actual talent and good/bad luck.

Unfortunately, I do not think this is a novel thought as it pertains to the betting markets. The fact is that the lines on recent games have reflected the Yankees’ true talents (actually, maybe more so since there are many Yankee fans around the nation and they may think, “they will turn it around soon, let’s bet them). And the betting lines will likely to continue to value the Yankees close to the same as they have before (which means slightly overrating them). The Yankees bad start probably will not generate inefficiency in the market.

1 comment:

results disoriented said...

While a lot of the Yankees' deviation from their expected records is due to variance, some is also because the back end of the bullpen (which was very good from 2004-06) is not performing well this year. Since these pitchers are put in at the most critical points, their results are amplified in the team's record.

If you look at teams like the Cubs, Reds, or last year's Indians, all of whom have fallen short of their expected W-L, you see many poor performances by the closers and set-up men, which account for a lot of the deviation. Here in Chicago, there are calls to trade Matt Murton for a reliever, even though they've spent over $40 million on the 'pen already.