That’s would be interesting but it’s not what I am doing. I am investigating the test method, i.e. does an increase in 200 pace predict an increase in long distance pace.
Hey Kevin, as others have said, just run an R^2 analysis. Also, for what its worth, you might want to read a couple papers on the usefulness of inferential statistics… for example, Hopkins, Cole, and Mason, 1998. The long and short of it (and people will want to argue) is that inferential procedures don’t tell you anything you don’t already know from visual inspection of your data.
Since I know everyone is intensely interested in the results 
here is the link.
It is very ugly, badly worded and needs lots of everything but the gist is there, if you want to track the change in your swim split performance but don’t want to actually swim 1650 yards, then you can probably use a 500 swim time trial to do so. If your 500 gets faster by 5%, you can be confident that your swim split is faster by about 5%. Seems stupid to say, however if you were doing 200s then you couldn’t be as sure. As for a 100 time trial, some of the folks in the class had faster distance performance when their 100 was slower.
If you’re doing time trials to track performance as input into a Banister model like in Phil Skiba’s software, then the unfortunate news from this is that you probably need to do a 500 instead of a 200 as your performance test.
If you’re doing time trials to track performance as input into a Banister model like in Phil Skiba’s software, then the unfortunate news from this is that you probably need to do a 500 instead of a 200 as your performance test.
No surprise there…