My favorite reader comment
I'm sorry that the reader comments don't post as fast as they used to, and hope we can find a way to rectify that situation, but I thought I'd highlight my favorite post from the last few hours. It comes from Harry, who felt my guarantee that the Orioles would win today was irresponsible and may have put Canadians in danger.
Harry's take: That's crazy talk---think of the consequences---an Orioles win would spark another Canadian riot---hell, I might even turn over my car and burn it...
My take: What was I thinking? Well, at least they have national health care.
One more thing: To those of you who wrote in to correct my math, you are correct about the real statistical probability of a coin coming up heads on the 17th consecutive flip. Not a tough concept. Where you come up short is recognizing my keen sense of humor, but maybe that's my fault.






> 
Comments
Actually, Pete, the O's 0-for-16 outcome can be used to dispute the hypothesis that the probability of the Orioles winning a game at Rogers Centre is 1/2 (a la "a coin toss").
For hypothesis testing, we assume that the probability of an O's win is 1/2, and then show how the statistical evidence (0-for-16) would be extremely unlikely under this hypothesis. Consequently we have to accept the alternate hypothesis that the probability of an O's win there is less than 1/2.
Here goes:
Assuming that the probability of an O's win at Rogers Centre is 1/2, what's the likelihood that the O's will win 0 games out of 16 there?
Prob (O's win 0 out of 16) = 0.00001526.
So this is pretty strong statistical proof that the O's do not have a 1/2 chance of winning a game at Rogers Centre (this assumes a number of conditions not worth mentioning here, but you get the drift).
In fact, we could continue this hypothesis testing assuming lower probabilities of winning:
Hypothesis: Prob of O's win at Rogers Centre is 4/10.
Prob (O's win 0 out of 16) = 0.000282.
So here again we have to accept the alternate hypothesis that Prob of O's win there is less than 4/10.
Sad to say, but continuing down this road will put the probability of an O's win there at 2/10 (at best):
Hypothesis: Prob of O's win at Rogers Centre is 2/10.
Prob (O's win 0 out of 16) = 0.02815, which isn't very strong but is not impossible assuming a 2/10 per-game win probability.
Thus it's safe to say that the O's don't have an equal chance of winning a game against the Blue Jays at Rogers Centre (for the time being).
Posted by: Lefty Fields | June 16, 2011 11:57 AM
Count me off the Nick Markakis bandwagon. I love the guy, but he has become an automatic out and should be hitting 9th. I'm not even exaggerating. Nick should volunteer himself to go down to the minors and work on his swing. He's hurting the team.
I'm listening to the game on the radio.They're making this rookie look like he is going to be a future HoF'er. I thought he had a 4.34 ERA in Double A. Are you kidding me? Do these guys have any shame? Did the kid even break a sweat through 2 innings?
Posted by: Mesotheliangelos | June 16, 2011 1:11 PM
Pete -
Doesn't ANYBODY on this %^&$(*& team know how to run the bases? No wonder they don't win!
Posted by: dave taylor | June 16, 2011 1:35 PM
Lefty Fields,
Your theory has more holes than a pasta strainer. Baseball game probabilities are affected by a multitude of actual and abstract statistical variables, in contrast with the results (heads or tails) or a coin flip.
As a crude example: a baseball game can be suspended. Shouldn't that be something accounted for in your probability model?
You're taking a sample size of 16 games over a span of 2+ years, during which period the O's may have had a total of 100 different players (each with different stats) on their roster. Then you're comparing the probability of those games to a coin flip?!
Posted by: Mesotheliangelos | June 16, 2011 1:45 PM
Meso,
Ease up. Obviously baseball involves more variables than my exercise could take into account, but I was trying to make the point that it's wrong to assume that the probability of the O's winning at Rogers Centre is 1/2. Pete was (facetiously) using the coin flip model in discussing the O's chances of breaking their Canadian losing streak.
However, I do think my exercise makes another point: either Rogers Centre is a helluva homefield advantage or the O's are not even up to a head-to-head challenge with an opponent that will "probably" end up in 4th place in the AL East.
Posted by: Lefty Fields | June 16, 2011 3:13 PM
This article is novel,Cheap Soccer Jerseys I really love it. I also have some ideas written in my article, you can have a look if you have the interest. didier drogba jerseysWe can also discuss different topics.
It is composed by sophsoccerjerseysmallcom 06.17.2011
Posted by: soccerjerseys | June 17, 2011 4:19 AM