Predicting Attendance

by torchy on Apr 21, 2008 10:34 AM EDT Comment 9 comments

We all love NAHWAL but I'm not so sure there is a away to predict points. Now, you can maximize points on players to pick (your best strategy is picking the hot bat/arm or picking someone consistently throughout the year) and by choosing sufficiently vague above and beyond picks (Escobar will make more putouts than errors). But the amazing thing is that your strategic picks mean about as much as a random guess. Yes, attendance is random. You can argue the distribution is centered a certain unknown mean and that choosing this mean based on past information (that's how the teams themselves do it by the way) is your best strategy. I propose a radically different idea. Choose the outliers. Choose "Rain Out" or "Sell Out". While you won't even be close on many days, these ceilings/floors must be accurate eventually and that's better than getting the close pick randomly (162 games/75 posts per game = about 2 points a year).

So here's your winning strategy:

1) Rain Out/Sellout

2) Today's Starter/This Week's Hot Bat

3) Vague reference to Gondeee's Mother, I mean vague pick

Good LUCK!

0 recs | Comment 9 comments

Story-email Email Printer Print

Comments

Acutally

The strategy for rain out is not really a good one, since no points are awarded for rained out games.

As for the vagueness of AAB picks, I reserve the right NOT to give you a point (and haven’t in the past, even this year) for things like “Braves win a game”—that is neither above nor beyond. As a matter of course, I also don’t like things that I can’t check against a box score. So if you say, “Kotsay scratches his balls while making a fab catch in the sixth,” even if he may have done so you probably won’t get the point since there’s no real way for me to check it if, for instance, I happened to be getting a beer at the very moment of the ball-scratch-catch.

But other than that, good post about NAHWAL strategy.

by gondeee on Apr 21, 2008 3:48 PM EDT reply actions 0 recs

I would imagine...

...that would have to make the top 10 webgems of the week.

by Smoltz's Beard on Apr 21, 2008 3:57 PM EDT up reply actions 0 recs

Good to know

Good info there Gondeee…I was wondering about that, when people would give way-too-easy AAB picks.

by VegasAces on Apr 21, 2008 5:46 PM EDT up reply actions 0 recs

WTF?

I was going to use the ol’ ball-scratch-catch for tomorrow’s AAB! Guess I’ll have to go with Diaz and the butt-sniff-whiff.

So, let me get this straight. NAHWAL is totally subjective based on 1.) gondeee’s sobriety, 2.) whether gondeee likes you, and 3.) gondeee’s motivation to check the facts.

Now I know, and knowing is half to battle!

by jug on Apr 21, 2008 6:51 PM EDT up reply actions 0 recs

Haha...

...good stuff.

by Smoltz's Beard on Apr 21, 2008 7:18 PM EDT up reply actions 0 recs

Thought you’d appreciate it.

by jug on Apr 21, 2008 7:57 PM EDT up reply actions 0 recs

You know...

Even if picking “rainout” garners no points (which i disagree with as you have a weekly prize and won’t be giving out points based on the other two categories) it’s still a valid strategy. I’m not saying that there exists a NE for any day, but if you pick based on highest probability, rainout and soldout should almost always be best responses. If you want the ABSOLUTE best strategy you can think about this whole game as a median voter strategy (see economics literature). The best pick is always the median attendance for the year when there are a small (even) number of votes. However, as we have many predictions the best strategy is to pick either one above or one below the highest/lowest pick so far. I chose not to offer this idea earlier most because it’s cheep and I’d prefer not to put it into some jackass’s mind but it IS the best strategy.

As for “vague” picks, I agree with you gondeee. Predicting anything so vague as a win is dumb. Remember, any pick in this category is at gondees’s whim for points. AAB points are rare so you must balance between the vague pick and the extraordinary pick.

NB: It turns out that as the number of pickers gets larger the strategy above collapses to the one a mentioned before. Just so you know…

Burning Brave

by torchy on Apr 22, 2008 1:08 PM EDT up reply actions 0 recs