Sniffing longshots thru variance.

[Stillriledup]

I was reading an article about NFL coaches and how the coaches of the teams who are the underdogs need to take many more chances in the games and not wait till the very end. Most coaches of underdogs just try to stay in the game and hang around and try and get lucky with a big play in the dying minutes, but the strategy should be to try and upset the "ebb and flow" of a normal game.

So, i was thinking that the racing equivalent of this would be a race that doesn't seem to have a normal flow to it.....maybe a wild speed horse who can't be rated...that would "mess up" a normal race and there might be more opportunity for something unforseen to happen. In a normal race where horses just line up and the pace is "honest' its more likely that the best horses will come home in the top 3. If you have a 20-1 shot going a 44 half and opening up 10 lengths, the race might "Fall apart" and you can get the odd result that you're looking for.

Here's the article on the NFL coaches.

http://www.advancedfootballanalytics.com/2009/05/are-nfl-coaches-too-timid.html

[classhandicapper]

That kind of thing makes sense in basketball.

If you are the inferior team you want to slow it down as much as possible (fewer possessions means fewer opportunities for the "long term" to exert itself) and you want to shoot and force as many 3s as possible to add variance.

I once read an article about this kind of thing in horse racing that was surprisingly written by David Sklansky of poker fame. I can't remember the details but it was geared towards the difference between a horse's chances of winning vs. his chances of running well and how that might create value in vertical exotics.

Horse A's last 10 races are 75 65 72 102 54 63 74 101 74 67

Horse B's last 10 races are 95 93 97 96 92 94 95 91 96 94

The chances of horse B winning and running well are way higher than A, but if A fires his best race he's very likely to win.