from Amazon
Thursday, June 16, 2011
Relationship of performance by days elapsed
A reader (Pat) did some research, and I was surprised by the results. He did great work.
So, I had to repeat his steps and expand the sample size. I looked at a performance of a player today to his performance tomorrow (one day elapsed) and tracked the change in his performance (via wOBA). I then look at his performance two days later (two days elapsed) and tracked the change in his performance. All the way down the line. This is based on 1993 to 2010 data. For the one day elapsed, I had 472,511 samples.
The number of the y-axis the difference is change in wOBA, and the number of the x-axis is the elapsed days. The regression equation and correlation is on the chart.
Now, while the pattern is clear, the scale on the left shows we’re not talking about a big deal. One day elapsed, and the change in wOBA is 0.285 points per PA. 181 days elapsed and it’s 0.291.
Not to mention that there’s “Seasonal” issues to take care of. If 181 days elapsed, we are talking about an April game compared to a late Sept, early Oct game. We might expect a larger difference simply for things like weather.
In any case, the chart itself is interesting to look at.
UPDATE: This is between two years. Once again, we see the same kind of pattern. The mean in both charts is an identical 0.287.
UPDATE: This is all comparisons for same year, next year, or 2 years out.
Recent comments
Older comments
Page 1 of 344 pages 1 2 3 > Last »Complete Archive – By Category
Complete Archive – By Date