Team USA North East Regional March 12-13, 2011

[Aaron Laing]

Aaron, can you explain what you mean by  first and second category?

I'll try (see below for the caveats)...

Category 1: If want a direct correlation between results and performance, you would normally want to control for those variables outside of a competitor's control--ie the random beat draw. If we look at the overall results, any competitor that drew a productive beat would naturally have the results skewed in his or her favour, so by removing the productive beats from the analysis you should give a better indication of performance vis-a-vis the results. I just picked what looked like the top three producers that did well outside of the good beats. You could of course rank everyone along that basis, but it's really the top couple that are significant finishes.

Category 2: The second "category" looked at the opposite side of the equation, looking at who performed well in particularly crappy beats. Now the problem with this "sample" is that not everyone fished those crappy beats, but of those that did some managed to pull out a fish or more when the bulk of participants didn't (or couldn't). I sort of look at it as the "iron chef" challenge--give a group of cooks a soggy eggplant to work with and the best chefs are able to do something with it. In the end you don't win a prize, but you get bragging rights.

Keep in mind this "analysis" is far from scientific. If you wanted to really come up with a proper statistical model you would need to factor in things like the beat splits, coin toss, time of day, etc. However, even if you did factor in all these variables (and a host of others) the final results wouldn't be that solid because the sample size is so small--200-300 competitors would make the results statistically relevant.

I was just trying to tease out a sense of who performed well under difficult conditions given the information at hand. When I compete looking at the overall results in this way gives me a means of examining my own performance, strengths and weakness, and comparing that against other competitors individually or in aggregate.

...course it could just be a way to sooth my battered ego after I place poorly

Aaron

*Edit: I just read Dejon's post on the method used to fish the beats--i.e. 28 half beats as opposed to 14 shared beats. I assumed shared beats, so my "analysis" based on the table has less statistical significance. However, if the information for the half beats was available the analysis would probably be stronger.

[Loren Williams]

Thanks Aaron.  I stunk at stats in college so what I concern myself with how I did on my beats compared to the other anglers.  Frankly, this time around I am disappointed with myself.  That said, conditions were so dynamic I'm not sure who to judge myself aside from I know I worked hard and doubt I should have done much different with the water I had.

[Aaron Laing]

I stunk at stats in college so what I concern myself with how I did on my beats compared to the other anglers.

Fair enough. I generally do the same, but this method gives a broad view of the results.

While many would argue against it, some sort of handicapping scheme based on beat results would actually give a better picture of performance than the current system. However, as I said to be really useful you would need a larger pool of competitors in any given event.

Oh well... idle musings when I should be working.

Aaron