flashlock
Banned Troll
Pure hogwash. Please provide a source of information that is not characterized by selection bias (see below).
What would be even better is if individuals posting statistics actually understood what "selection bias" is....
There is a best-selling business book entitled, "From Good to Great" that is a great example of selection bias.
The author of this book took a bunch of then "successful firms" (GE, Xerox, etc.), looked at what they had in common, and then wrote a book saying, "If you do what they do and have X (e.g., a dedicated M&A division, international operations, skunks works, separated CEO-Chairman positions, and a thousand other things), you will be successful too!"
The problem is that he didn't "control for" all of the firms that also had X but FAILED. Moreover, he didn't "control for" all the firms that elected *NOT* to have X in the first place, but succeeded.
That means that there are two types of "selection bias." The former is known as "survival bias" because you are only looking at firms that have X *AND* are successful. The latter is known as "self-selection bias" because whether or not a firm chooses to engage in a behavior (i.e., X) affects the results if you are only looking at those who choose to engage in such behavior.
Precisely. This is "self-selection bias"--cops will grab because they have to subdue the guy....
Similarly, it is not surprising that grunts will grab, and then report back to their superiors that the confrontation involved grappling. If a grunt has his weapon, he's going to shoot it. The only reason he is grappling is that is what he was trained to do.
I doubt highly that some dude in a robe in Iraq is going to willingly tackle a U.S. grunt wearing 100+ pounds of gear. So it is the grunts who are tackling people. Why? Because they are trained to do that....
If you can't see how statistics need to be interpreted in light of various caveats, then I guess the conversation stops there...and your statistics need to be flushed down the toilet.
Sorry, dude.
Bujinkan Budo Taijutsu is, at base, at weapons art. Even a child will pick up something and want to use it as a weapon to give itself an advantage. Why not learn to use things around you to advantage? You've based your arguments on supposed "logic". It seems pretty logical to me to want to learn how to use things around you to advantage and would prefer to keep that advantage even in a "fake experiment" of 50 people in a crowded place or open field....
Wow! What a great post!!!
I don't know who you are, meta asethete, but you are a very wise creature..... Thanks for posting!
-ben
I have not seen the raw data, and I don't know if it's possible to do so--maybe it's classified. The reports, from my understanding of them, were not regarding police type actions (I must tackle the Afghani and not shoot him so I can question him), but rather getting ambushed, the gun jams, and you're fighting. The US Army has shifted to BJJ as its core, and even trains its special forces in these methods now.
So the US military is putting its soldier's lives on the line to promote Gracie JJ? They biased the data, at the risk of lives, for that? Maybe I should be more cynical... would you be if they were training everyone in Taijutsu? Maybe then it wouldn't seem so biased...
I stand by the material: the soldiers were given detailed questions about specifics during their conflicts, and they discovered that 100% of the fights had elements of grappling (most grappling mixed with striking); NONE had striking alone. 0. Nada. Zilch.
Where's the bias? I don't know how many out here would agree, but in very full on sparring, things naturally go to grappling when one striker starts striking better--at least I've experienced that (been on both ends). You don't have to believe in those stats, but can you show evidence where real life or death altercations had only elements of striking? I'll even let you sway the results in your favor for your bias... cite one source that shows that, and I'll agree with you.
Otherwise, you should take back your comments.