Friday, June 08, 2007

trend spotting update

another weekly installment of the trending update finds the cubs and brewers tracking one another much as they have for most of this year. past postings are may 11, may 14 and june 1.



the cubs rebounded for their cyclical low on game 53 and experienced a positive crossover/convergence at game 55, evidence of a positive trend that has in the past forecast a winning run of baseball over the following 10-20 games. from the outset of the year to the cyclical high on game 31, the club went 16-15; from the high to the low, the club went 6-16; since the low, the cubs have won four of six.



unfortunately, despite losing two of three in their recent series against the cubs, the division-leading milwaukee brewers have also experienced a positive crossover/convergence as of their last game (game 60), which follows on their cyclical extreme of game 48. this would portend better times for the brewers as well, who have rebounded from the low in rather dull fashion. from the outset to the cyclical high at game 36, the brewers went 25-11; from the high to the low, 3-9; and from the low to today, 5-7.



to this we add the saint louis cardinals, who have surpassed the cubs for the moment to reach second place in the division. do they present a continuing threat?

it hardly seems so on this analysis. inversely to the pattern being carved out by the cubs and brewers, the cardinals struck first a cyclical macd low in game 30 of (-2.04), at which point their record sat at 12-18. the subsequent reversal in trend raced to a cyclical high at game 40, in which span the club went 4-6. the crossover had appeared in game 36, however, and true to form the positive trend in 12-game pythagorean record persisted through game 52 -- some 16 games. but it is now on the wane as the 12- and 26-game lines converge. one would expect on the principles of the analysis for the next event to be a negative crossover, and for the uptrending cubs probably then to surpass the cardinals.

of further interest -- it was earlier noted in the post that first set this page to examining trends in runs scored and allowed that run differential as a means of forecasting final records was most often a victim of insufficient sample size until well into the season.

a simple analysis would be simply to look back at several seasons and try to divine at what point in a normal season year-to-date run differential -- and the pythagorean estimated record that might be derived from it -- first approach the final figure. though perhaps inelegant, such an analysis is easy -- and the answer in examining the last four seasons (2003-6) suggests that run differential becomes material at a point approaching or in excess of 60 games, though even afterward the data can drift.


major league ballclubs have now come to that point of the season where consideration of these factors has become sensible. and what are the tallies?







the reason not to concern ourselves too much about the cardinals becomes even clearer here than previously -- while the cardinal offense has surged to life over the last month, its pitching remains in difficult straits and shows little sign thusfar of remediating. one might expect the cardinals to play better going forward than they have overall to date, but even a break-even record between today's 26-31 and the end seems as a prediction something of an eyebrow-raiser.

but what is most encouraging is the brewers' chart. this page had previously written at milwaukee's highwatermark:

in the hard light of close examination, dear reader, it seems far more likely that, rather than closing the gap, the cubs will continue to bleed games to milwaukee in fits and starts for the remainder of the season as they have to date. not only have the brewers posted a much larger run differential and greater theoretical record thusfar, not only have they done so on far less anomalous measures of luck, but they've actually won games doing it on a strength of schedule that is essentially no different from the cubs'.


but more data has changed the picture, as it sometimes does. instead, given what we now know, it would seem as likely for the brewers to bleed back a few games over the remainder of the season as it would the cubs to. were one to base opinion solely on this graph, indeed it would seem likely that the cubs would close the gap at least partially before year end; however, one must also consider the good fortune of cub pitching in terms of babip -- a feature of their outperformance to date which has subsided since the start of may and seems likely to continue to revert over the remainder of the year much as milwaukee's has continued to improve toward the league mean.

No comments: