by maligned » Tue Jan 24, 2006 6:59 pm
I think you're right, Dave, despite what I wrote. TSN has always claimed that the choice to run is not based on exact thresholds. My only point was that it seems like HAL considers the final raw numerical probability, rather than just stealing someone a certain percentage of times.
Basically, there are two ways to look at how HAL may consider running someone. Let’s look at a scenario: Torii Hunter has a 13 steal rating when he doesn’t get his good lead. Imagine that I’ve set my strategy settings for stealing to “aggressive.” Now, the two ways we could look at how HAL decides whether to run Hunter or now would be the following: 1) HAL considers the fact that Hunter is supposed to be aggressive, and steals when he doesn’t have his good lead a certain percentage of the time, based on the 13 rating, no matter what the hold and arm ratings of the pitcher and catcher may be, or 2) HAL looks at the final adjusted rating (plus or minus the hold and arm ratings of the pitcher and catcher), THEN steals a certain percentage of the time based on that number. Imagine the hold rating is –2 and the catcher’s arm is –1. Now the final numerical chance is actually 10, instead of 13.
My original point was that I think HAL considers this final numerical chance, NOT the original rating of 13 when deciding whether to steal a certain percentage of times. So, I think once you have your settings in place, HAL will steal a certain percentage of times on a 13 final probability and a certain percentage of times on a 17 probability…no matter who the basestealer is. It seems like adjusting your aggression or lack thereof against certain teams isn’t as important as it seems on the surface. If you’re set to aggressive, HAL will go a higher percentage of the time on a 15 probability than he would on a normal setting. Your changing of settings only seems to affect how HAL decides to act once he has the final probability in hand.
As you said in your post, Dave, when HAL decides to go is still unpredictable and not an exact science…I just think we might be looking at his decision-making process wrong at times.