Quote:
Originally Posted by AltonKelsey
Not sure I understand the problem. This is not rocket science.
If I want to know how a trainer does with first time starters , I look at all his first time starters .
If I want to see how gate breakthroughs perform, I take a list exactly like the one provided and analyze that .
If they win at half the expected rate , I have information.
I don't think I need a database consultant to figure this out
I.
|
Who said you need a data consultant? No, it's not rocket science. It's research science.
If you want to know how gate breakers perform you compare them to other starters in general. You need a control group to get an impact value and expected rate.
If you want the best way to see how this trainer does with 1st time starters you want to know how they do compared to his other starters or the population of 1st timers in general. You need a control group to get an impact value an and expected rate. In this case you might be able to get away with the "quick and dirty" method and eliminate a control group though I would prefer to see just how his win%, etc compares to others.
Just do what Quirin did. It's simple stats. there's a reason this methodology has developed.
All I can do is tell you what any statistician would. In the end it's your choice and your money.
Good luck, whatever you do.