Thursday, April 12, 2012

The Streak

In Will’s post this week he mentions that fans should not be too upset with the Yanks going 0 and 3 to start the year. He is right about that. Baseball is, like any other team sport, very streaky. Teams can go on a tear one week and fall the other way the next. Not just teams, but players as well. (Note that, just now, Jose Bautista is in a slump). It will happen and they will get out of it. After several years a player can have an odd better year and a worse year but most years will balance themselves out and, after around four years in the bigs, the stat machine is pretty well set. For the most part they are what they are.

So I got to thinking about when do we know a team is good, bad or ugly? The point I am trying to make is that while streaks do happen (and often) it appears a team’s fortune for the year is established early on, and over the course of several games. The percentage after 10 and or twenty games is close to the way most teams fair for that year. Most contending teams do follow this weird kind of record.

So as curious as it may seem, here are some interesting looks (not exhaustive) at some successful years and some not so much. I will look at the first ten and twenty games then compare to the end of the season:

Jays. The Back2Back years. In 1992 won their first six and went 9/1 for ten and 12/8 (.600) for the twenty and ended the year with 96/66 (.593). In 1993 they went 6/4, 12/8 and ended with 96/66 (.586). In 1985, again, it was very similar. They went 6/4, 13/7, and ended 99/62 (.615). That loss to the Royals in ALCS was the worst in Jay’s history.

Tigers. In 2006 the Tigers won their first 5. But using my sequencing they really went 6/4, 12/8 and ended 95/67 for a years record of .586. It was the same percentage for the whole year. In 1968 it was 9/1, 13/7 and ended with 103/5 (.650). 1984 was the exception with 35/5 (.875) and ending 104/58 (.641).

Yankees. As Will said, those 1998 Yankees did go 0-3. But for the first ten games they went 6/4 and at 20 games went 15/5 (.750). Pretty impressive recovery, I’d say. (This same no name team lost a series to the Jays in August the same year). By end of the year they 114/48 for a whopping .704 on the season and won the WS as well. However, the 1927 (the big name team) Yankees had the best season record ever and still went 7/3 (.700), 13/7 (.650) and ended with 110/44 for an even more whopping .714.

Cardinals. The 2011 wild Cards went 4/6, 11/9 (.555) and ended with 90/72 (.556) and also the WS ring to wear. Go figure. In the heydays of 1942- 44, ending with St Louis Showdown and such names as Stan “The Man” Musial, Johnny Hopp and Mort Cooper, went the way of all the above, with small variation went 6/4,11/9 and ended in the .680s.

At the other end of the chart are the 2011 Jays for example. They went 5/5, 9/11 and ended 81/81. Another little quirk is that at 40 games in they were 20/20. The 2011 Pirates went 5/5, 9/11 (.450), At 40 games they were still 18/22 (.450) and ended the season at 72/90 (.444). The 2003 Tigers went 1/9, 2/18 and 9/31 to end 43/119 (.265). It is fairly consistent with this weird approach.

This is all very strange indeed. With looks at other teams, it appears that for the most part the way a team starts the season is the way it finishes. Maybe Sabermetrics should look at this as well. I certainly do not pretend to understand all of Sabermetrics or it’s variations. Bill James, I am sure, can come up with a formula to predict the year’s outcomes. It would smooth out the streaks and account for the usual injuries and other anomalies that occur every year to each team. Let me know what you think. Is this ridiculous or is there something to this approach. Or is it only individual player statistics that count? Help me refine this search. I want to know.

Thanks, Will. I would not have looked into this without your post.

1 comment:

Larry Toman said...

Hey John, a very interesting approach indeed. I knew you would eventually fall into the sabermetrics fold, to a certain degree. I think you should find out how to contact Bill James, and run this by him. I would add one more item to your unique approach. Take what you said, and then make one more observation at the All-Star break. May or may not be able to draw one more correlation. Have a great weekend, and cheers to Jack.